International Journal of Computer Applications in Technology (65 papers in press)
Non-linear modified equation modeling in dynamical systems (Case Study research on Long Jump patterns)
by Farzad Sharifat
Improving Arabic Text Categorization using FA Words with K-Nearest Neighbor and Centroid-Based classification algorithms
by El-Sayed Atlam, M.E. Abd El-Monsef, O. El-Barbary
3D Scanning Machine and Additive Manufacturing: Concurrent Product and Process Development
by Ismet P. Ilyas
Simulation and visualisation approach for accidents in chemical plants
by Feng Ting-Fan, Tan Jing, Liu Jin, Deng Wensheng
Abstract: A new general approach to lay the foundation for building a more effective and real-time evacuation system for accidents in chemical plants is presented. In this work, we build the mathematical models and realise automatic grid generating based on the physical models stored in advance with several algorithms in jMonkeyEngine environment. Meanwhile, the results of the simulation data through finite difference method (FDM) are visualised coupling with the physical models. Taking fire as an example, including fire with single and multiple ignition sources, shows the feasibility of the presented approach. Furthermore, a coarse alarm and evacuation system from fire have been developed with a multiple SceneNode and roam system, which also includes the making and importing of the physical models. However, to improve the accuracy of the mathematical models, adaptability and refinement of the grids and universality of the evacuation system is the direction of efforts.
Keywords: simulation; chemical accidents; alarm and evacuation system; jMonkeyEngine.
Detecting occluded faces in unconstrained crowd digital pictures
by Chandana Withana, S. Janahiram, Abeer Alsadoon, A.M.S. Rahma
Abstract: Face detection and recognition mechanisms, a concept known as face detection, are widely used in various multimedia and security devices. There are significant numbers of studies into face recognition, particularly for image processing and computer vision. However, there remain significant challenges in existing systems owing to limitations behind algorithms. Viola Jones and Cascade Classifier are considered the best algorithms from existing systems. They can detect faces in an unconstrained crowd scene with half and full face detection methods. However, limitations of these systems are affecting accuracy and processing time. This project proposes a solution called Viola Jones and Cascade (VJaC), based on the study of current systems, features and limitations. This system considered three main factors: processing time, accuracy and training. These factors are tested on different sample images, and compared with current systems.
Keywords: face detection; unconstrained crowd digital pictures; face recognition.
Adaptive self-recurrent wavelet neural network and sliding mode controller/observer for a slider crank mechanism
by Ahmad Taher Azar, Fernando E. Serrano, Josep M. Rossell, Sundarapandian Vaidyanathan, Quanmin Zhu
Abstract: In this paper, a novel control strategy based on an adaptive self-recurrent wavelet neural network (SRWNN) and a sliding mode controller/observer for a slider crank mechanism is proposed. The aim is to reduce the tracking error of the linear displacement of this mechanism while following a specified trajectory. The controller design consists of two parts. The first one is a sliding mode control strategy and the second part is an SRWNN controller. This controller is trained offline first, and then the SRWNN weights are updated online by the adaptive control law. Apart from the hybrid control strategy proposed in this paper, a velocity observer is implemented to replace the use of velocity sensors. The outcomes obtained in the numerical experiment section prove that the smallest tracking error is obtained for the linear and angular displacements in comparison with other strategies found in the literature owing to the uncertainty and disturbance rejection properties of the sliding mode and the self-recurrent wavelet neural network controllers. Besides, it corroborates an accurate velocity estimation due to the precise theoretical design of the proposed observer in order to reduce the error between the measured and estimated state to zero. This paper begins with the derivation of the dynamic equations of the slider crank mechanism followed by the derivation of the proposed observer and control strategies. Finally, to demonstrate the effectiveness of the controller/observer strategy, a numerical example is supplied to analyse the variables of the system, tracking error and the estimated variables
Keywords: adaptive wavelet neural networks; sliding mode control; sliding mode observer; slider crank mechanism.
A runtime model-based framework for specifying and verifying adaptive RTE systems
by Nissaf Fredj, Yessine Hadj Kacem, Mohamed Abid
Abstract: Adaptive Real-Time Embedded Systems (RTES) may execute in an unpredictable context that is impossible to definitely consider in the development time. Therefore, these systems are required to adapt their response to unpredicted changes at runtime in order to maintain their feasibility and usefulness. Their design requires effective runtime modelling formalisms for monitoring, reconfiguration planning and adaptive system analysis. In this context, software designers need to evaluate, refine and validate runtime models at early stages of development via adaptation tools, in particular for runtime adaptive systems to avoid execution problems. In the present paper, we propose a runtime model-based framework that allows the modelling and simulation as well as the traceability of adaptive RTES. Our proposal starts by a high-level specification based on the UML/MARTE profile, which describes an adaptive system and supports the reasoning about its behaviour and structure at runtime. The runtime UML/MARTE models are translated into an adaptive one that instantiates MAPE patterns for the control and the traceability of the runtime system. Then Model-to-Text (M2T) transformations allow us to generate simulation scripts for the analysis of adaptive system behaviour at runtime and evaluate its real-time constraints.
Keywords: RTES; runtime adaptation; runtime model; M2T transformation; MAPE control loop.
Skew decision process based on machine learning content analysis and clustering
by Tanzila Saba
Abstract: Skew decision process includes skew detection and correction that is mandatory for automatic information retrieval from documents. Hence, there is an utmost need to correct the information skew prior to further processing. Accordingly, this research presents a robust document information skew detection and correction approach based on histogram clustering for efficient information retrieval. The proposed approach is quite generic and, therefore, could detect skew angle for various types of document such as graphics, charts, postal labels, handwritten text, forms, drawings and their possible combination. The proposed approach is robust that could deal with up to
Keywords: skew decision process; automatic information processing; document analysis; semantic web.
Simulation results and practical implementation of a PD-super-twisting second order sliding mode tracking control for a differential wheeled mobile robot
by Ebrahim Elyoussef, Nardênio Martins, Douglas Bertol, Edson De Pieri, Ubirajara Moreno
Abstract: A robust solution to the trajectory tracking control problem for a differential wheeled mobile robot should deal with the existence of parametric and structural uncertainties, external disturbances and operation limitations. The first order sliding mode control with boundary layer is a common and suitable solution that can ensure chattering attenuation, but with poor degree of robustness. Fortunately, higher order sliding mode control can achieves greater degree of robustness with the reduction of the chattering phenomenon. Based on this knowledge, a control strategy is proposed using a super-twisting sliding mode control, which enforces a second order sliding mode, integrated with a proportional plus derivative control to solve the problem achieving good robustness. This linear control technique plays an important role in increasing the robustness by mitigating the influence of neglected dynamics. Simulation and experimental results are explored to prove the effectiveness of the proposed control strategy.
Keywords: differential wheeled mobile robot; trajectory tracking; sliding mode control; PD control; chattering attenuation; uncertainties and disturbances.
Identification of nonlinear stochastic systems using a new Hammerstein-Wiener neural network: a simulation study through nonlinear hydraulic process
by Saif Eddine Abouda, Donia Ben Halima Abid, Mourad Elloumi, Yassine Koubaa, Abdessattar Chaari
Abstract: Hammerstein-Wiener models have been proved to be suitable in modelling a class of typical nonlinear dynamic systems. This paper aims at developing a Hammerstein-Wiener Neural Network (HWNN) which formulates Hammerstein-Wiener mathematical model, in order to identify a nonlinear dynamic system operating in stochastic environment. A central aspect is that a general situation has been considered wherein non-invertible nonlinearity output and correlation of stochastic disturbances after the dynamic linear block. Different from the existing parameter identification methods, the model is developed to handle two types of learning algorithm that can directly obtain the parameters of the unknown time-varying nonlinear system. Firstly, all neural network weights in HWNN are adapted using a Back Propagation based Gradient algorithm (BPG). The second type, namely Recursive Least Square Back Propagation based Gradient method (RLSBPG), is derived from the BPG algorithm to achieve the parametric estimation of Hammerstein scheme where the remaining parameters are estimated by the least-squares approach based on fuzzy technique to ameliorate the estimation quality. The convergence analysis of the algorithms is presented, and their performances are tested through a simulation study of a nonlinear hydraulic process.
Keywords: nonlinear stochastic systems; Hammerstein-Wiener mathematical model; Hammerstein-Wiener neural network; BPG learning algorithm; RLSBPG learning algorithm; fuzzy technique; convergence analysis; hydraulic process.
Fast recognition and classification of static and dynamic signs for Persian Sign Language
by Milad Moghaddam, Manoochehr Nahvi, Negin Pourmomtaz
Abstract: Sign language is the most effective way for communication between deaf and hearing-impaired people. Since most non-deaf people are not familiar with sign language, a vision-based translator/interpreter can be a very useful tool to enhance their communication. This paper presents a recognition system for Persian static and dynamic signs. The system is designed based on proposed modified non-linear kernel-based fast feature extraction methods, consisting of hybrid kernel principal component analysis and hybrid kernel discriminant analysis. For recognition of dynamic signs, the proposed feature extraction method is employed in association with spatio-temporal approach. The proposed methods are examined and compared with several existing feature extraction methods, including linear and non-linear kernel-based methods. The experiments indicate that our feature extraction methods significantly outperform other methods and reduce computational time while they achieve high recognition rates. Our simulations achieved a promising classification accuracy rate of 96.78% and 96.99% on static and dynamic signs, respectively.
Keywords: human-computer interaction; Persian sign language recognition; hand gestures; kernel-based feature extraction; static signs; dynamic signs.
A parallel multiobjective swarm intelligence framework for big data analysis
by Amr AbdelAziz, Kareem A. Ghany, Taysir Soliman, Adel Sewisy
Abstract: Nowadays, data generated from smart devices, such as sensors, computers, and tablets in huge volumes, with different formats, and at a high pace, comply with big data characteristics. Big data led to the emergence of new technologies, such as Hadoop and Spark. They provide both management and analysis of big data. Analysing big data is a time-consuming process when using traditional data mining techniques. Swarm Intelligence (SI) are population-based metaheuristic methods inspired by the behaviour of bird flocks in nature. Particle swarm and ant colony optimisation are examples of these methods. They have been combined with data mining techniques to solve MultiObjective Problems (MOPs) in small and medium-sized data, presenting good performance. However, when applying SI methods to solve MOPs in big data, an efficient scalable framework will be required, such as MapReduce. MapReduce is a programming framework developed to execute tasks in parallel. In this paper, we summarise new technologies proposed to manage and analyse big data. We present how metaheuristics can be adapted with big data technologies. We characterise problems that arise when analysing big data MOPs, in addition to proposing methods to overcome these problems, giving examples in bioinformatics field.
Keywords: big data; big data analysis; data mining; particle swarm optimisation; multiobjective optimisation; MapReduce; Spark.
Age identification of Chinese rice wine using electronic nose
by Wei Ding, Peiyi Zhu, Ya Gu
Abstract: This paper is concerned with the identification of the age of Chinese rice wine. To address this problem, a new electronic nose system with the multivariate analysis method based on the artificial olfactory technique is developed. First, four features are extracted to represent the dynamic behaviour of the signal that is generated from the array of the Taguchi Gas
Keywords: age identification; Chinese rice wine; electronic nose system; multivariate analysis.
Effective packet multiplexing Method to improve bandwidth use
by Mosleh Abualhaj, Qusai Shambour, AbdelRahman Hussein
Abstract: Voice over IP (VoIP) technology is suffering from squandering the network bandwidth because of the very hight packet header overhead. Packet multiplexing methods are used to reduce packet header overhead. In this paper, a new packet multiplexing method, called Voice Frame Pruning and Multiplexing (PM-VFrm), is proposed to improve VoIP bandwidth use. The PM-VFrm achieves this in two aspects. First, by multiplexing several packets that share the same route in one header. Unlike the current methods, the PM-VFrm will not add any additional header to the VoIP packets, because it reemploys some of the fields in the RTP header. Second, reducing the VoIP packets payload size by pruning the voice frame based on a specific algorithm. Again, none of the current methods has proposed pruning the voice frame. The PM-VFrm method outperforms comparable methods. The saved bandwidth has exceeded 55% compared with the traditional method (without multiplexing).
Keywords: VoIP; RTP protocol; VoIP bandwidth use; VoIP packets multiplexing.
Neural network control and performance simulation of an active control mount with an oscillating coil actuator
by Rang-Lin Fan, Jia-Ao Chen, Zhen-Nan Fei, Chu-Yuan Zhang, Fang-Hua Yao, Quan-Fa Wu
Abstract: Active control mounts (ACMs) are an effective solution to improve the comfort of passenger cars in terms of acoustics and vibration. Neural networks are widely used because of their powerful learning capabilities and their ability to approximate arbitrary nonlinear functions. This paper aims to apply neural networks to ACMs and explores the general process of neural network ACMs. A three-layer BP neural network model (NNM) with a 6-8-1 structure based on a sigmoid activation function is established with an oscillating coil actuator (OCA) as the controlled object. The actuator output force is collected as training and test samples when it is excited under different types of input current signals, such as single-frequency constant amplitude, swept-frequency constant amplitude, and random currents. Learning and testing are performed for the NNM, and the results show that the identified NNM based on random signals has good accuracy and a strong generalization ability. Based on this well-identified NNM, two control methods neural network direct self-tuning control and NNM reference control and their implementations are discussed; the active control simulations are performed in MATLAB Simulink. The simulation results for typical low, medium and high frequencies show that both control methods achieve good vibration isolation effects. This research shows the strong adaptability of neural networks, which lays a good foundation for subsequent control system development and experimental verification as well as research on active neural network control for mounting systems.
Keywords: neural network; active control; active control mount; oscillating coil actuator; lumped parameter model; system identification; powertrain mounting system; automotive.
Discrete stochastic modelling of computer viruses' prevalence on a reduced scale-free network
by Mohamed Essouifi, Abdelfattah Achahbar
Abstract: We construct a stochastic version of the deterministic epidemiological model, introduced in Yang and Yang (2014), to study computer viruses spreading across a network with two degrees. We compare the results expected by these two approaches. To be closer to real situations, where randomness is always present, we use the discrete time Markov chain method to explore the dynamic behaviour of infection and susceptibility densities of nodes in the network. Monte Carlo simulations, in good agreement with the deterministic approach especially when the proportion N1/N2 increases, lead to the conclusion that any attempt to entirely eradicate network viruses would prove unavailing. As a consequence, the only alternative choice is to contain their prevalence by adopting countermeasures. Our stochastic modelling approach is systematic and shows a good performance in predicting the dynamic behaviour of the computer virus propagation.
Keywords: epidemiological model; computer viruses; network with two degrees; discrete time Markov chain; Monte Carlo simulations; stochastic modelling.
Multi-level analysis of IEC 61131-3 languages to detect clones
by Jnanamurthy HK, Raoul Jetley, Frans Henskens, David Paul, Mark Wallis, S.D. Sudarsan
Abstract: Nowadays, automation is gaining more popularity due to its potential to complete tasks more quickly and reduce the need for human intervention. Automation can be assisted by using programmable logic controllers (PLCs). To improve the development times of applications and execution time of PLCs, software re-usability and optimisation has an important role in the production. PLCs are typically programmed with IEC 61131-3 languages to automate and implement the applications. PLC program classification plays an important role in the identification of similar functionality, which can be considered as software clones. There are many existing methods to identify software clones, but these methods mainly focus on structure-based analysis to detect clones and not detect clones in domain-specific IEC 61131-3 languages. In this paper, we present work to identify clones in IEC 61131-3 languages, using an approach based on four different perspectives: 1) clone prediction: filtering based on heuristics to reduce program comparisons; 2) structural analysis: detect syntactic code clones which are present in program organisation unit; 3) semantic analysis: analysis of output variable dependency and input variable impact usage to detect semantic clones; 4) variable interval analysis: analysis of each variable intervals which pass through the control flow graph (CFG) to collect intervals and use it to examine and detect clones. We analysed IEC 61131-3 language structured text programs to detect clones. Our approach is a combination of structural, semantic and data interval based analysis. As a result, our approach is feasible and yields good result in detecting clones on our test data.
Keywords: structural clones; semantic clones; variable interval analysis; controller programs; software maintenance.
The use of higher-degree polynomials for trajectory planning with jerk, acceleration and velocity constraints
by Marek Boryga
Abstract: The paper presents the smooth trajectory planning method using higher-degree polynomials. Polynomial functions of acceleration were determined and accurate dependencies were derived for the calculation of polynomial coefficients and motion time in the case of velocity, acceleration and jerk constraints. The paper also presents a method for trajectory planning when a few constraints occur at the same time. To assess the efficiency of the proposed approach, a simulation study has been carried out using MATLAB. The courses of kinematic quantities are presented in the diagrams and compared. The proposed method ensures the continuity of displacement, velocity, acceleration and jerk and can be applied for trajectory planning of various mechanical systems (manipulators, mobile robots, CNC machines, autonomous vehicles).
Keywords: trajectory planning; smooth trajectory; higher-degree polynomial; kinematic constraints; mechanical system.
The construction and implementation of the positive-negative pressure pulse control system for oil well plug removal
by Zhenfu Ma, Yuankai Song, Kai Zhang
Abstract: Owing to oilfields in high water cut stage and multi-set well patterns parallel exploiting, exploitation and underground conditions has become much more complex and problems such as oil well mineralisation blockage and scaling are obvious, seriously impacting the oilfield water flooding efficiency. This paper designs a positive-negative pressure pulse control system for oil well plug removal based on the principle of positive and negative pulse, implements the main equipment selection, PLC and software programming, constructs a set of plug removal control system, and successfully conducts tests of the positive-negative pressure pulse control system for oil well plug removal.
Keywords: oil well plug removal; positive-negative pressure pulse; PLC; Kingview.
A survey for recent applications and variants of nature-inspired Immune Search Algorithm
by Faisal Alkhateeb, Ra'ed M. Al-Khatib, Iyad Abu Doush
Abstract: Artificial immune system (AIS) is a well-known nature-inspired and population-based algorithm that proved its effectiveness for solving engineering and practical real-world problems. AIS can adapt to learning, has many models for different immune systems, which can be used to tackle different kinds of optimisation problems, and it can also be hybridised with other algorithms. In this paper, we extensively summarise the recent researches of AIS and categorise them based on the application problem to understand the current trend of the usage of this algorithm. In addition, we provide the up-to-date open research problems that are not solved by immune search algorithm, and they were solved recently by other algorithms. This can help in paving the road for future research directions in the AIS field.
Keywords: artificial intelligence; immune algorithm; nature-inspired metaheuristics; optimisations; applications; variants of immune algorithm.
Parameter optimization and experimental verification of the positive-negative pressure pulse automatic control system for plug removal
by Zhenfu Ma, Yuankai Song, Kai Zhang
Abstract: The effective application of water injection wells in oilfield operations plays an important role in improving the production quality of the oilfield and the actual revenue of oilfield enterprises. Aiming at the current problems of mineralisation blockage and scaling in oil production, this paper designs a positive-negative pressure pulse automatic control system for plug removal. First, it studies the main equipment for the implementation of positive-negative pulse automatic control, and then designs the technical principles, control logic and implementation methods of positive pressure pulse, negative pressure pulse, and positive-negative coupling control. Next, the control effect is verified by the automatic control experiment platform and the control parameters are optimised. Finally, the optimised control effect is verified by experiment using the control platform.
Keywords: oil well scaling and plugging; positive-negative pressure pulse; automatic control; Kingview.
Research on rotating high frequency current injection method based on LC resonance network
by Li-ping Zhong
Abstract: The high frequency signal injection method is an effective method to realise the sensorless detection of the rotor position of the permanent magnet synchronous motor (PMSM), which can estimate any position of the rotor in the whole speed range. When the frequency of the injected signal is high, the rotating high frequency current injection method can obtain more significant signal with rotor position information for PMSM with small salient characteristics, so it can obtain higher position estimation accuracy. However, the traditional method of injecting high-frequency current needs a current source inverter or a complex control mode, which is difficult to apply in practice. The drive circuit based on LC resonant network can convert the resonant voltage signal into the current signal, thus the driving voltage and rotating high-frequency current signals can be simultaneously provided to the PMSM through the voltage source inverter, which greatly facilitates the extraction and processing of rotor position information and improves the sensorless detection accuracy of rotor position. The experimental results verified the effectiveness of the circuit and method.
Keywords: LC resonant network; rotating high frequency current injection; high frequency negative sequence component; double PI regulator.
An improved 2-OPT optimisation scheme for Hamilton loops
by Bo Sun, Shicai Liu, Yongquan You, Chuanxiang Ren
Abstract: Aiming at the problem that the traditional traffic route planning has a single target strategy and cannot be adjusted according to the actual situation, a multi-lane planning model based on Hamilton loops is proposed, which uses the characteristics of Hamilton loops to cover all nodes in the set and uses the nearest neighbour algorithm to obtain the initial loop. The OPT algorithm optimises the initial loop, avoids the problem that the 2-OPT algorithm cannot find the optimal loop due to randomness, and improves the efficiency of the 2-OPT algorithm. Considering road congestion and other situations, the design introduces speed parameters and comprehensive influence factors to obtain different choices and obtain the optimal solution of the Hamilton loop in different situations, which proves the effectiveness and feasibility of the algorithm. The optimised model can give different path planning schemes according to different actual situations and can find the shortest time path and the shortest path, making the Hamilton loop model more practical. The model uses the adjacency matrix to visually represent city nodes. The simulation results verify the introduction of variable parameters, such as speed influence factors, to meet different planning needs, achieve more targeted route planning, improve the path planning scheme, and improve people's travel efficiency.
Keywords: Hamilton loop; 2-OPT algorithm; path planning.
Automatically optimised stereoscopic camera control based on an assessment of 3D video quality of experience
by Dawei Lu, Xiaoguang Huang, Zhi Li, Zhao Zhang
Abstract: Stereoscopic 3D (S3D) technologies have gained significant attention owing to their widely applications. However, producing high-quality S3D content is still a challenging task that requires careful handling to achieve the artistic intent and maintain visual comfort. In this study, we present an automatically controlled stereoscopic camera controller that specifically addresses the challenges in S3D content production. The key idea that distinguishes our method from the existing work is that our method aims to predict the 3D quality of experience (QoE) in the production stage so that the optimised camera parameters can be obtained automatically. To this end, considering two interconstraint indicators, i.e., visual comfort and perceived depth, we collect and label a dataset of S3D video scene clips and generate a 3D video QoE assessment model that can guide the optimisation of the stereoscopic camera parameters. We describe how to implement our system into a modern production pipeline that has been used in some projects, including commercial ones. The experimental results, including the user studies, demonstrate that our system enhances the perceived depth without creating visual fatigue and that our controller can make the production of S3D content easier and more efficient.
Keywords: stereoscopic 3D; camera control; visual comfort; perceived depth; 3D quality of experience.
Feature vector sharing and scale comprehensive optimisation for target detection in smart neighbourhood governance and monitoring
by Jianmin Liu
Abstract: This article proposes feature vector sharing and a scale comprehensive optimisation strategy of image target detection and recognition method of complex street maximum suppression based on the calculation of the corresponding feature area corresponding to the feature map and complete eigenvector. Based on this, this article also combines a fine-tuning method based on transfer learning generalisation, which is suitable for non-convex optimisation and high-dimensional space. First, the method described above implements the optimal rectangular selection box competition based on the scale comprehensive optimisation strategy, and selects the selection box that can reflect the core essence of the target in each classification set. Then, this article realises the model of detecting an image target in a complex neighbourhood, which improves the accuracy and robustness. Furthermore, we experimentally demonstrate that the accuracy and robustness of our proposed method are superior to those of conventional methods.
Keywords: feature vector sharing; scale comprehensive optimisation; neighbourhood images; smart neighbourhood monitoring.
An identity-based integrity verification scheme for cloud storage in 5G environment
by Zuodong Wu, Jianwei Zhang, Zengyu Cai
Abstract: In order to solve the security problems such as leakage and tampering of sensitive data in cloud storage, the previous schemes usually sacrificed communication efficiency for higher security, which caused serious computing overhead. Therefore, in this paper, we adopt the idea of Chinese Commercial Cryptographic SM9 and SM3 algorithms, and regard user identity identification as interference factor and verification mechanism, so as to propose an identity-based integrity verification scheme for cloud storage. Furthermore, we also give the security analysis under the assumption of Diffie-Hellman and discrete logarithm problem on elliptic curve. Finally, simulation results show that our scheme can not only verify the integrity of sensitive data correctly, but also resist common malicious attacks. Especially in terms of efficiency, our scheme can effectively reduce the storage and computing burden, space and time cost. This will have a certain guiding significance for the privacy protection of cloud storage in 5G environment.
Keywords: cloud storage; Chinese Commercial Cryptographic; SM9; SM3; identification; Diffie-Hellman; discrete logarithm; integrity; privacy protection; 5G.
An accessibility learning system for higher integrated education of hearing-impaired students' technology
by Jian Zhao, Liu Wang, Lijuan Shi, Zhejun Kuang, Di Zhao
Abstract: Integrated education for the disabled is the embodiment of social equity and social progress, and China has carried out pilot work of integrated higher education for deaf students for four years. However,in the actual teaching process, it is found that the learning efficiency and effect of deaf students are quite different from that of healthy students. This difference is mainly caused by hearing impairment. Hearing-impaired students are accustomed to understanding the semantic meaning mainly through lip language. However,the amount of information of lip language of hearing-impaired students decreases according to distance, shielding and other interference reasons, which leads to obstacles to the semantic understanding of course teaching. An accessibility learning system for higher integrated education of hearing-impaired students is established to help them to capture the auxiliary information of teachers lip language and voice. It is also evaluated to explore its practical application effect. The evaluation results show that this system can help hearing-impaired students to better understand the semantic meaning of the course and integrate into the integrated education classroom.
Keywords: hearing-impaired students; integrated education; accessibility learning Ssystem.
Measurement of sentence similarity based on constituency parsing and dilated convolution
by MingYu Ji, ChenLong Wang, Gang Liu
Abstract: Measurement of sentence similarity is widely used in the field of natural language processing, the current mainstream method is based on neural network similarity model. In actual application, the method of neural network has some disadvantages. On the one hand, when sentences are input into the neural network, the problem of semantic loss is caused by the interception and zero-filling operation of the sentence that is too long or too short. On the other hand, it ignores the semantic relation between interval words. Thus, this paper proposes a method of measuring sentence similarity based on constituency parsing and dilated convolution, by using constituency parsing to design rules to reduce unimportant semantic components of long sentences and to supplement important semantic components of short sentences. In addition, the receptive fields in sentence dimension and word vector dimension are dilated to capture the semantic association of the two-dimensional interval words. Finally, the method is verified on two datasets.
Keywords: sentence similarity; neural network; constituency parsing; semantic component; dilated convolution.
Hybrid image denoising based on region division
by Yong Tian, Jing Wang, Yunfeng Zhang
Abstract: This paper proposes an efficient hybrid framework for image denoising, in which the advantages of different denoising methods are effectively incorporated by using the region prior knowledge. In detail, the input noisy image is first divided into a large number of overlapping patches followed by extraction of speed-up robust feature (SURF), and then the noisy patches are classified into two categories based on twin support vector machine (TWSVM). The texture patches are enhanced via gradient histogram preservation (GHP) while flat patches can be re-established using Block Matching Three-Dimensional Filtering (BM3D). Finally, the re-established images can be acquired by fusing the processing results of the two kinds of patch. To evaluate the effectiveness of the presented method, we conduct experiments on standard image datasets and compare the performance with other outstanding denoising approaches. Experimental results show that the presented method achieves better results, especially in containing textures and edges compared with existing image denoising methods.
Keywords: image denoising; hybrid framework; twin support vector machine; gradient histogram preservation; block matching three-dimensional filtering.
Predictive analytics for spam email classification using machine learning techniques
by Pradeep Kumar
Abstract: Automated text classification is the most widely used approach to manage an enormous amount of unstructured text data in digital forms, which is continuously increasing across the globe. Machine learning techniques are applied for automatic email filtering effectively to detect the spam mail and prevent them from delivering into the user's inbox. This paper used logistic regression, k-nearest neighbors, naive Bayes, decision trees, AdaBoost, ANNs, and SVMs for spam email classification. All the classifiers are learned, and the performance is measured in terms of precision, recall, and accuracy using a set of systematic experiments conducted on the Spambase dataset extracted from the UCI Machine Learning Repository. The effectiveness of each model was empirically illustrated to find a better and viable alternative model. The quantitative performance analysis of supervised and hybrid learning techniques is presented in detail. Experimental results indicate that ensemble methods outperform in terms of accuracy compared with other methods applied.
Keywords: text analytics; feature selection; predictive modelling; spam filtering; machine learning techniques.
Mechanical analysis and simulation of colliding damage of castor capsule
by Junming Hou, Yong Yang, Jingbo Bai, Hongjie Zhu
Abstract: Mechanical damage in the process of castor capsule shelling is the main problem for castors. In the paper, the effects of collision speed, moisture, collision material curvature and thickness on the collision deformation and equivalent stress of castor capsule are discussed. Based on the Hertz collision theory and dynamic equations, we established a hybrid collision mechanics model of castor capsule, and used finite element method on the simulation of its collision and shelling process. The results show that as the collision continues, the velocity and acceleration of the collision are nonlinear. Based on the optimal space filling (OSF) and multi-objective genetic algorithm (MOGA), the related parameters are determined. The shelling process model is established, and the damage of castor was studied. According to the response surface optimisation (RSO) method, the optimal parameter ratio of the castor capsule shell breaking was obtained. The results were that the hull part curvature was 340 mm, the shelling part thickness was 1.6 mm, the castor capsule capsule elastic modulus was 46.03 MPa and the collision speed was 6.96 m/s. The maximum equivalent stress at this time was 0.28 MPa, and the total deformation was 28.52 mm. The paper can provide reference for the optimal design of castor capsule shelling equipment.
Keywords: castor capsule; collision damage; simulation; genetic algorithm; response surface optimisation.
Design and implementation of a real-time digital signal processing system using PIC24 microcontroller and wireless GUI control
by Shensheng Tang, Alex Stangl, Manish Ale Magar, Shubham K C
Abstract: In this paper, a real-time digital signal processing (DSP) system is designed and implemented by using a PIC24 microcontroller circuit and a C# GUI application running on PC. The wireless communication between the PIC24 subsystem and the GUI subsystem is implemented via Bluetooth modules on the subsystems. The DSP system first digitises an input square signal of a certain frequency through an on-chip ADC of the PIC24 microcontroller, then uses different FIR digital filters to extract certain harmonics of the input signal, and outputs it as a sinusoidal signal to an on-chip DAC as well as sending the sampled data and filtered data over Bluetooth to the GUI. The GUI, besides plotting the input and output waveforms, can provide a means of controlling all functionalities of the system through a developed communication protocol. The design and implementation for the proposed DSP system are successfully demonstrated by experimental results. The hardware and software co-design method can be extended to other industrial applications and used as a good paradigm of engineering education for college students.
Keywords: digital signal processing (DSP); PIC24 microcontroller; ADC; DAC; C programming; C# programming; GUI; FIR filter; Bluetooth.
Design of MIMO QFT fractional control based on intelligent fractional PID? controller combined with decentralised and centralised FBLFD prefilter: application to SCARA robot
by Asma Aribi, Najah Yousfi, Nabil Derbel
Abstract: The aim of this paper is to propose new robust controls for multivariable parametric uncertain systems and to validate their efficiency in robot path tracking. An automated fractional multivariable quantitative feedback theory is developed. The principle is to obtain desired performances on the basis of controllers and prefilters without using a loop-shaping process. The proposed approach has benefited from the robustness of fractional control. Indeed, the method is based on a combination of an intelligent fractional PID? controller with both diagonal and non-diagonal frequency band limited fractional differentiator. A bi-objective optimisation based on genetic algorithm is used to find the controller parameters. The developed methodologies are applied to a SCARA robot model and the findings highlight the robustness of the designed controller and the success of the diagonal prefilter to eliminate loop interactions.
Keywords: FBLFD prefilter; fractional controller; bi-objective optimisation; multivariable systems; path tracking; quantitative feedback theory; SCARA robot.
FPGA design and circuit implementation of a new four-dimensional multistable hyperchaotic system with coexisting attractors
by Sundarapandian Vaidyanathan, Esteban Tlelo-Cuautle, Aceng Sambas, Leutcho Gervais Dolvis, Omar Guillén-Fernández
Abstract: This research work focuses upon the FPGA design and electronic circuit implementation of a new multistable four-dimensional hyperchaotic system. This work starts with the dynamics and phase plots of a new four-dimensional hyperchaos system. A detailed bifurcation analysis is carried out for the new system, and special properties such as multistability with coexisting attractors are reported for the system. An electronic circuit model using MultiSim of the new hyperchaos system is designed. Finally, an FPGA-based design of the new system is performed by applying two numerical methods. Circuit simulation and FPGA design of the new hyperchaos system are very useful for practical applications of the system.
Keywords: hyperchaos; hyperchaotic system; multistability; circuit simulation; FPGA design.
Connecting historical changes for cross-version software defect prediction
by Xue Bai, Hua Zhou, Hongji Yang, Dong Wang
Abstract: In the whole software life cycle, software defects are inevitable and increase the cost of software development and evolution. Cross-Version Software Defect Prediction (CVSDP) aims at learning the defect patterns from the historical data of previous software versions to distinguish buggy software modules from clean ones. In CVSDP, metrics are intrinsic properties associated with the external manifestation of defects. However, traditional software defect measures ignore the sequential information of changes during software evolution process which may play a crucial role in CVSDP. Therefore, researchers tried to connect traditional metrics across versions as a new kind of evolution metrics. This study proposed a new way to connect historical sequence of metrics based on change sequence named HCSM and designed a novel deep learning algorithm GDNN as a classifier to process it. Compared to the traditional metrics approaches and other relevant approaches, the proposed approach fits in projects with stable and orderly defect control trend.
Keywords: software testing; cross-version defect prediction; software metrics; historical change sequences; deep learning; deep neural networks; gate recurrent unit.
Hierarchical Smart Routing Protocol for Wireless Sensor Networks
by Amine Kardi, Rachid Zagrouba
Abstract: This paper proposes a new hierarchical routing protocol dedicated to Wireless Sensor Networks. Several solutions have been proposed, but an optimal one, which can be applied in different situations, still remains nonexistent. Our proposal takes advantage of existing solutions and proposed a new radial cluster head (CHs) selection algorithm to ensure the QoS, a good load balancing and enhance the network lifetime. A performance analysis of our proposal with LEACH, Mod_LEACH and M-GEAR protocols is presented in this paper comparing these protocols using different metrics such as network lifetime, throughput, stability and instability periods, remaining energy, network overhead etc.
The simulations show that our solution outperforms existing solutions due to the radial architecture proposed for CHs selection and the use of two transmission levels for intra and inter-cluster communication. Obtained results show that our proposal improves the network lifetime by 113% compared to LEACH protocol and optimizes the energy consumption by more than 62% in comparison to M-GEAR.
Keywords: Routing protocol; Wireless Sensor Networks; Hierarchical; Network; Cluster Head
Real Time Sign Languages Character Recognition
by Sari Awwad, Sahar Idwan, Hasan Gharaibeh
Abstract: Deaf people face many challenges to communicate with the hearing world. Many studies and industrial solutions came up with interpretation from sign language to normal text but most of them are limited either to static images for letters or static animated character plays the word in motion. Therefore, this research worked on enhancing the feasible algorithms been used like image detection, image processing techniques and image translation methods.
The contribution consists of two parts, the first one is using our dataset, and the second one is proposing a new solution by extracting surf features after using image filtering based on the existing methods to accelerate the translation process for the long sentences.
Experimental results over our dataset report a distinguished accuracy compared with other studies in terms of efficiency time and recognition rate of sign language character recognition.
Keywords: Sign Language Dataset; Image filtering; Surf Features; M-SVM2; Sign CharacterrnRecognition.
Special Issue on: Computational Intelligence and Applications
Intelligent game-based learning: an effective learning model approach
by Tanzila Saba
Abstract: Game-Based Learning (GBL) broadly refers to the use of video games applications to support teaching and learning processes. This research focuses on the concept of GBL in the context of stimulating interest in the field of computer science education specifically. In contrast to theoretical learning, GBL is a practical learning approach that is meant to teach and be enjoyed at the same time. Additionally, a GBL model with visual features has been proposed and tested. Promising feedback has received from learners through the post conducted surveys. The research findings exhibit that GBL is an effective methodology in transferring knowledge, enhancing learning, and making the learning a more enjoyable process in computer science studies than just the theoretical approach.
Keywords: binary games; game-based learning; logical games; theoretical learning.
Special Issue on: Modelling and Expert Intelligence for Traditional Chinese Medicine Diagnosis and Knowledge Base Development
Effects of low frequency somatosensory music on heart rate and skin temperature in healthy people
by Baohong Mi, Lixin Ren, Jialin Song
Abstract: With the rapid development of biology, brain science, psychoanalysis, etc., the biological effects of low frequency somatosensory music on the human body are gradually being discovered. However, its objective evaluation system is not yet sound. In this paper, heart rate and skin temperature of healthy people are monitored by heart rate acquisition equipment and medical infrared thermal imager in different modes of low frequency somatosensory music. Results show that there was a significant difference in heart rate between experimental groups and control group (P < 0.05), and the skin temperature also had significant difference between experimental groups and control group (P < 0.01), but there was no significant difference in music frequency change and skin temperature change (P > 0.05). It is concluded that low frequency somatosensory music can stimulate parasympathetic nerve and change the heart rate of healthy people, and it also can be used to change the metabolic state of the human body by regulating mental activities, and then reduce skin temperature. This paper proposes a new objective evaluation method for low frequency somatosensory music therapy.
Keywords: low frequency somatosensory music; infrared thermal imaging; heart rate; skin temperature.
Knowledge discovery for spleen yang deficiency syndrome based on attribute partial order structure diagram
by Hui Meng, Xiaoying Han
Abstract: Syndromes and medications in traditional Chinese medicine (TCM) have been studied by advanced information technology in the modernisation of TCM, promoting the development of knowledge discovery in TCM. In this paper, based on the method of attribute partial order structure diagram (APOSD), syndrome-symptom APOSD and prescription-herb APOSD are constructed for spleen yang deficiency syndrome, which is a common syndrome in TCM. The common symptoms and specific symptoms of spleen yang deficiency syndrome are extracted from the syndrome-symptom APOSD. The association rules among herbs in the prescriptions for treating spleen yang deficiency syndrome are visualised on the prescription-herb APOSD, and herb pairs and herb groups are extracted. Compared with Apriori algorithm, APOSD not only obtains important association rules, but also shows the compositions of prescriptions in the diagram. APOSD provides a new scientific research method for TCM.
Keywords: attribute partial order structure diagram; knowledge discovery; spleen yang deficiency syndrome; symptom patterns; prescription compatibility.
Application research on quantitative prediction of TCM syndrome differentiation based on ensemble learning
by Huaixin Liang, Xin Yang, Shaoxiong Li, Siheng Chen, Xiaoqing Zhang
Abstract: A quantitative prediction method for TCM syndrome element identification based on ensemble learning is proposed. Four comparative experiments were designed. Firstly, eight mainstream learners were used to perform the regression prediction based on the symptoms and syndrome values using the quantitative data of clinical TCM syndrome differentiation. Secondly, five learners with excellent prediction performance were selected to design three integrated learners including homogeneous static integrated learner, heterogeneous static integrated learner and dynamic one, where the heterogeneous integrator used as the learner weight coefficient to weigh up its significance. By comparing the MAE and MSE of the three ensemble learning methods in the four syndrome differentiation groups, it is found that the regression effect based on heterogeneous ensemble learning is the best (MAE: 0.012; MSE:4.55E-04: 0.733), and the principal sequential evaluation of syndrome elements gained relatively matching degree, which proves the feasibility of the application of the method proposed in the quantitative prediction of clinical TCM syndromes.
Keywords: ensemble learning; syndrome differentiation; regression prediction; multi-objective learning; traditional Chinese medicine; syndrome element model; quantitative prediction; artificial intelligence; clinical auxiliary diagnosis; homogeneous static ensemble learning; heterogeneous dynamically integrated weighted regression; heterogeneous static ensemble learning.
Formal concept attribute partial-order structure diagram and applications
by Yunli Ren
Abstract: Formal concept attribute partial-order structure diagram (FC-APOSD) is proposed on the basis of formal concept analysis and attribute partial-order structure diagram (APOSD) theory, which inherits the advantage of good visualisation of APOSD and can reveal relationships among formal concepts at the same time. This paper focuses on the mathematical description of FC-APOSD, how to use FC-APOSD to explore knowledge from a formal context, and the applications of FC-APOSD. Specifically, our work mainly includes the following points: (1) FC-APOSD is described in mathematical formal language; (2) after giving a necessary and sufficient condition for an attribute set of a node to be the intension of a formal concept, a method to find pseudo-intensions using FC-APOSD is provided; (3) the relationship among objects is discussed using groups and subgroups in FC-APOSD; (4) a method to find exclusive attributes from a formal context is provided by using FC-APOSD; (5) with the help of classical Chinese medicinal formulae related to emotional diseases from Treatise on Febrile and Miscellaneous Diseases, the syndrome-treatment pattern of emotional diseases is explored using the FC-APOSD method.
Keywords: formal concept analysis; attribute partial-order structure diagram; knowledge discovery; syndrome-treatment pattern.
Statistical analysis for User Group of Opposing Traditional Chinese Medicine in Weibo
by Li Hao, Zhang Bingzhu, An Xuzhao, Ma Xingguang, Shen Junhui
Abstract: With Western culture and science being widely accepted in China, Traditional Chinese Medicine (TCM) becomes a controversy. Therefore, it is very important to study the public's views on TCM. The rapid development of online social networks, such as Sina Weibo, can quickly and easily provide a sample for emotional analysis. In this research, firstly, Sina Weibo's users of TCM were collected, and their blogs were obtained. The blogs were automatically marked as supporting TCM or opposing TCM based on user tags. Then, blogs about TCM were selected by using the Chinese word segmentation tool and TCM dictionary. Finally, the Chinese word segmentation tool was used to count the words and find the highest frequency of words in the blogs of opposing TCM to explore hot topics and propose corresponding suggestions for the development of TCM.
Keywords: emotional analysis; Chinese word segmentation tool; traditional Chinese medicine; Weibo; user tag.
Research on the regular pattern of Professor Saimei Li using Chinese medicine alone in treating middle-aged type 2 diabetes mellitus on the basis of partial order structure theory
by Xu Sunjing, Hao Yizhao, Li Yat Tung, Li Saimei
Abstract: The method of knowledge discovery based on partial order structure theory was first proposed by Professor Hong Wenxue of Yanshan University. In recent years, this method has been widely applied to the study of classical Traditional Chinese Medicine (TCM) and the inheritance of academic experience of famous Chinese medicine. It provides a new method to solve the problem of knowledge discovery and inheritance of Chinese medicine. In this research, the information of symptoms (prescriptions) of herbs from 50 effective cases of middle-aged type 2 diabetes mellitus (T2DM) treated by Professor Li Saimei with Chinese medicine was processed with formal contexts, and then the corresponding partially ordered diagrams were obtained for the exploration of the medication rules of Professor Saimei in treating T2DM. The results showed that the common symptoms of the middle-aged patients are mainly of heat syndrome - the main symptoms are string pulse, poor quality of sleep, yellow fur on the tongue, red tongue, and smooth pulse. In regard of the treatment, the main prescription formula is Lizhong Wan combined with Gegen Qin Liang Tang for middle-aged patients. The combination of warm and cold formulas aims at warming the spleen and dispelling the pathogenic wind, clearing pathogenic heat and eliminating dampness. This research can promote the inheritance and innovative development of Chinese medicine in an effective way.
Keywords: middle-aged type 2 diabetes mellitus; partial order structure theory; knowledge discovery; Li Saimei.
Diagnosing Parkinsons disease with speech signal based on convolutional neural network
by Tao Zhang, Yajuan Zhang, Yuyang Cao, Lin Li, Lianwang Hao
Abstract: Dysarthria is one of the typical early symptoms of Parkinsons Disease (PD), and that is the basis of diagnosing PD with the speech signal. In this paper, we propose a novel method to analyse the speech signal by Convolutional Neural Network (CNN). At first, the time series signal of speech is converted into spectrograms to represent the time and frequency features in a signal figure; and then, we train the CNN with the spectrograms and their labels from the training set. At last, we test the network precision by the test set of speech signals. The experiments show the accuracy of the method is 91%, which is outperforming the traditional classification for speech signals.
Keywords: diagnosing; Parkinson's disease; speech signal; convolutional neural network; dysarthria; spectrograms.
State Chinese medicine theory based on the mathematical description TCM principles and modelling of complex human system
by Jialin Song, Wenxue Hong, Cunguo Yu, Xiaoyun Wu, Jingbin Wang, Cunfang Zheng
Abstract: Based on the essence of traditional Chinese medicine, the theory of State Chinese Medicine (SCM) proposed and established profoundly reveals that traditional Chinese medicine is state medicine. SCM fully expresses the main characteristics of 'holistic concept' and 'syndrome differentiation and treatment' of TCM, which is the description of TCM theory under the vision of modern science. The establishment of SCM theory can make more non-TCM professionals understand TCM, master the core principles of TCM theory, and use modern science and technology to develop TCM to serve their own health and public health. It will promote the formalised mathematical description of the principles of traditional Chinese medicine, and promote the measurable, estimable, computable and quantifiable evaluation of human health states. It will be expected that a high degree of integration of modern science and technology with traditional Chinese medicine could be achieved, and make contributions to human health and the progress of traditional Chinese medicine.
Keywords: state Chinese medicine; computable model; modelling of complex human system; syndrome differentiation and treatment.
A novel classification tree based on local minimum Gini index and attribute partial order structure diagram
by Cunfang Zheng
Abstract: Decision tree is not only an important machine learning method, but also the basis of ensemble learning methods such as random forest and deep forest. Based on the theory of formal concept analysis (FCA) and attribute partial order structure diagram (APOSD), a new decision tree for classification is proposed in this paper. Firstly, the local minimum of Gini index is used to complete the data granulation, and the formal decision mode information table (FDMIT) is constructed. Then, the attribute partial order classification tree (APOCT) is generated based on APOSD to complete the pattern recognition and rule extraction. The method of APOCT separates the process of granulation and visualisation, and the granulation process is easy to parallelise and efficient. The experimental results show that APOCT is effective.
Keywords: classification tree; decision tree; partial order; Gini index; data granulation; formal concept analysis; knowledge discovery.
The physical pattern evaluation and identification method of infrared thermal image of human health state in traditional Chinese Medicine
by Wenxue Hong, Baohong Mi, Cunguo Yu, Wenzheng Zhang, Jingbin Wang, Cunfang Zheng
Abstract: In view of the lack of objective detection methods for human health status in traditional Chinese medicine (TCM), this paper proposes a method based on infrared thermal image and sign mode evaluation and identification. This method involves data collection, basic medical theory, basic methods, basic models and other core contents. From the perspective of infrared thermal image data collection, three thermal image classification methods, physiological thermal image, interference thermal image and health thermal image, are proposed. From the point of view of medical theory, the medical theory of infrared thermography is as follows: governing exterior to infer interior, recognise the whole through observation of the small parts, master both permanence and change. From the perspective of basic methods, health assessment is mainly based on the calculation of human symmetry, uniformity, thermal sequence and three-dimensional information. On the basis of establishing the model of 'depression, dampness, deficiency and stasis' in the evolution of human health state in TCM, this paper puts forward the technical route of evaluation and identification of infrared thermal image physical sign modes, which starts from light and shallow, grows from obvious, becomes from cold and hot, and finally comes to the pattern. At last, the common infrared thermal imaging modes in clinic are given. This paper is of great significance to establish an objective method for the evaluation and identification of infrared physical signs based on human health status of TCM.
Keywords: traditional Chinese medicine; infrared thermal image; evaluation and identification; human health state.
Assumption of constructing intelligent recommendation model of diabetic Chinese patent medicines
by Chaonan Liu, Yuzhou Liu, Enliang Yan, Jianfeng Fang
Abstract: Responding to the urgent requirements for rational use of diabetic TCM patent medicines, this study comprehensively, objectively and accurately reveals the 'treatment-formula-drug-dosage-property' knowledge of Chinese patent medicines on the basis of TCM principles, EBM, clinical practice and partial order theory. It establishes a multi-level, partial-order visualised expression method for TCM treatment in diabetes, and constructs an intelligent recommendation model of diabetic Chinese patent medicines, which provides technological approaches for promoting rational use of Chinese patent medicines. The completed results prove that this method could effectively find out practical guiding knowledge and give reasonable suggestions for drug use.
Keywords: diabetes; Chinese patent medicines; machine learning; knowledge discovery.
Special Issue on: Advanced Big Data and Artificial Intelligence Technologies for Edge Computing
An improved hybrid error control path tracking intelligent algorithm for omnidirectional AGV on ROS
by Yaqiu Liu, Hui Jing
Abstract: In order to improve the accuracy and stability of intelligent omnidirectional AGV path tracking based on mecanum wheels, an improved intelligent hybrid error control path tracking method is proposed. The method combines the angular velocity of the intelligent AGV vehicle with the error correction of longitudinal velocity as the coupling estimation error. The coupling estimation error and the improved pure tracking algorithm are combined as the lateral control of the intelligent AGV car, while the PID control is used as the vertical control to further reduce the error interference. The ROS simulation results showed that compared with the tracking effect of the traditional pure tracking algorithm, the tracking path of the improved intelligent hybrid error control path tracking algorithm is closer to the real path, which greatly improves the trajectory deviation phenomenon, and the path tracking accuracy and stability are significantly improved.
Keywords: mecanum wheel; path tracking; improved hybrid error control; coupling estimation error; intelligent omnidirectional AGV; pure pursuit; ROS.
Path discovery approach for mobile data gathering in wireless sensor networks
by A.N.U. Raj, Shiva Prakash
Abstract: Data gathering is one of the most important fundamental tasks in wireless sensor networks (WSNs). It consumes large amounts of energy of the sensor nodes, which reduces the network lifetime. So, mobile sinks or mobile collectors are often used to collect data from the sensor nodes in WSNs. But the main challenge is to discover the optimal path for a mobile sink that reduces the energy consumption of the sensor nodes. We propose a Modified Travelling Path Planning (MTPP) algorithm to find the shortest travelling path for a mobile sink that reduces the data latency as well as energy consumption of the sensor node. This method provides an effective data gathering mechanism for the mobile sink. In this method, the mobile sink has to traverse along the chord of the communication circle of the nodes. In this approach, the travelling path is the sum of chords of all the communication range and line segments between them. After that we apply a B-spline smoothing curve method over it. The effectiveness of proposed method is verified through mathematical proving and in MATLAB. This method is used to find the best possible path for a mobile sink or collector that will enhance the network lifetime as well as reduce the energy consumption of the sensor node.
Keywords: data gathering; wireless sensor network; mobile sink; network lifetime.
A system architecture for intelligent agriculture based on edge computing
by Li Liu, Qian Wang, Bo-qun Li
Abstract: Agricultural informatisation has been a major aspect in the field of agriculture owing to advancement in communication technologies. Many studies have been conducted to assist production, such as Wireless Sensor Network (WSN) and cloud computing. However, few studies have been focused on the severe requirements of delay and energy consumption of the mobile node. In order to improve network performance, edge computing is introduced into our system architecture for intelligent agriculture. Mobile Edge Computing (MEC) is a distributed computing architecture to offload some tasks to the edge of core network. We adopt a hierarchical structure made of three layers: physical perception layer, information service layer and intelligent application layer. In addition, we further confirm an offloading model for intelligent agriculture. In this article, the state-of-the-art development in intelligent agriculture and current approach of its key technology are discussed. The potential opportunities and challenges of the proposed architecture are presented as well.
Keywords: edge computing; WSN; mobile edge computing; intelligent agriculture.
Computer image analysis for various shading factors segmentation in forest canopy using convolutional neural networks
by Liangkuan Zhu, Jingyu Wang, Kexin Li
Abstract: Determination of the various parts in the forest canopy images is critical because it reflects a variety of parameters for plant population growth in forest ecosystems. Recently, deep learning has become one of the most promising techniques in machine learning for image analysis. However, only a few studies on applications in forestry information fields are published. This study presents the use of deep learning for the detection of various shading factors in hemispherical photographs of the forest canopy. First, a forest canopy hemispherical photographs dataset that can be used for the research of related algorithms is constructed. Based on FCN, the dilated convolution layers and multi-scale feature fusion are used to improve the accuracy of forest canopy image segmentation. Furthermore, the conditional random field (CRF) is adopted to optimise the results. The loss function is used to control the selection of different features of the model. Finally, experiments show that this method can achieve automatic segmentation of the sky, leaves, and trunks of forest canopy images. Compared with the FCN model, the average pixel accuracy of the improved FCN model is improved by 9.11%, and it has good robustness.
Keywords: forest canopy image; image segmentation; full convolutional neural network; dilated convolution; multi-feature fusion; conditional random field.
Special Issue on: Computational Advances in Healthcare Solutions
Performance analysis of surrounding cylindrical gate all around nanowire transistor for biomedical application
by Amit Agarwal, Prashanta Chandra Pradhan, Bibhu Prasad Swain
Abstract: Transistors have been used for more than a decade for fast and accurate detection in biomedical fields such as biomaterials sensing devices. There have been many researchers working on biosensor devices, but we are more focused on the deep micron transistor device, which exhibits high sensitivity and accurate results. This paper presents a highly sensitive, more accurate and faster device using a silicon on insulator based cylindrical surrounding gate all around nanowire (SCGAA-NW) transistor. This proposed device can be used for biomedical applications, e.g. diabetes sensor, gas sensor, pressure sensor and different substances present in the blood or environment by setting and analysing proper physical parameter of the device. In this paper, we have varied different physical parameters, i.e. channel material (i.e. SiN, CdS, GaN, ZnO, GaP, Si, GaAs, Ge), oxide material (i.e. SiO2, Si3N4, Al2O3, Er2O3, Y2O3, HfO2, Ta2O5, La2O3), channel radius (1-10 nm), oxide thickness (1-10 nm), concentration of different materials on the sensor acting as gate to source voltage, drain to source voltage (-0.5 V to 0.5 V), and channel doping (10^7 to 10^14) for the most suitable biomedical application in different environments. Analytical modelling of the SCGAA-NW transistor has been done by solving the 1-D Poisson equation, using Gauss law and parabolic approximation method. Also, we have shown the mathematical equation that relates to the impact on gate to source voltage owing to use of sensor material. In this paper, we have investigated that with changes in physical parameters of the device, there is impact on the potential at the channel surface. We have implemented the SCGAA-NW transistor model and plotted the potential at the channel surface with different device parameters graphs using Matlab Simulator.
Keywords: biosensor; biomedical application; gate all around nanowire; MOSFET; cylindrical channel; potential at the channel surface.
Fuzzy logic system for diabetic eye morbidity prediction
by Tejas V. Bhatt, Raksha K. Patel, Himal B. Chitara, Gonçalo Marques, Akash Kumar Bhoi
Abstract: Diabetes is a common chronic disease; the number of people who are affected by this health problem is increasing worldwide, leading to a high cost for healthcare systems. Therefore, the main contribution of this paper is to present a fuzzy logic system for diabetic eye morbidity prediction. This work is divided into two parts. The first part is the examination of eye vision by the ophthalmologist and also other examinations such as postprandial blood sugar, hypotension, cholesterol, and duration of diabetes. The second part is the analysis of 400 patients medical records collected in 2019. The fuzzy system proposed for prediction of diabetes retinopathy provides reliable accuracy for eye-vision threatening and eye morbidity. The proposed fuzzy system has five input parameters and one output parameter, which predicts diabetic neuropathy. The input parameters are random blood sugar, postprandial blood sugar, hypotension, cholesterol, and eye vision. The output parameter is the morbidity in diabetic retinopathies, which are non-proliferative, proliferative, and clinically significant macular edema. The proposed system is designed to support the endocrinologist and ophthalmologist in the diagnosis of diabetic retinopathy.
Keywords: artificial intelligence; blood sugar; cholesterol; diabetic retinopathy; fuzzy set theory.
Detection of bifurcations and crossover points from retinal vasculature map using modified windows feature-point detection Approach
by Meenu Garg, Sheifali Gupta, Soumya Ranjan Nayak
Abstract: Identification of feature points such as bifurcation points and crossover points in retinal fundus images is useful for predicting the various cardiovascular diseases. In this paper, a new approach called the Modified Window Feature-Point Detection (MWFD) is proposed to identify the vascular feature points in the fundus image. The MWFD technique makes use of two different windows, 3 x 3 and 5 x 5, with alternative vessel pixel property for the detection of all feature points. In this method, firstly, skeletonisation is done to obtain a one-pixel width vasculature map for the identification of feature points. Then spur removal operation is performed to reduce the error generated due to skeletonisation. After that, two windows with alternative vessel pixel property are applied on the skeletonised vasculature map to improve the identification of feature points. Also, adjacent feature points are removed by checking the 8-adjacency between them. This paper also resolves the problem of conversion of one crossover point into two bifurcation points generated due to skeletonisation. The simulation is performed on the DRIVE database using MATLAB software. Simulation results show quantitative improvements in detection of feature points from retinal fundus image by increasing the number of true positives and reducing false positives and false negatives. These results provide an efficient and reliable technique for analysis of various retinal structures.
Keywords: fundus image; retinal vasculature map; feature point; bifurcation point; crossover point; retinal detachment.
Survivability prediction of patients suffering hepatocellular carcinoma using diverse classifier ensemble of grafted decision tree
by Ranjit Panigrahi, Moumita Pramanik, Udit Kumar Chakraborty, Akash Kumar Bhoi
Abstract: The mortality rate of patients who have cancer is the second highest cause of death around the globe. Hepatocellular carcinoma (HCC), a type of liver cancer, is one such cause of death. Though the survivability probability of patients is very rare, a mechanism to predict chances of survival will provide a great aid to the medical practitioners who treat patients suffering from HCC. In this article, two state of the art survivability prediction schemes have been proposed separately for male and female subjects suffering HCC. The prediction engine employs Feature Selection Via concave minimisation (FSV) feature ranking and the Sigmis feature selection scheme to extract limited features of both male and female subjects, and an ensemble of decision tree grafting mechanism successfully predicts the chances of survivability of HCC patients. The gender-specific survivability prediction engine is the first-ever such prediction model for the diagnosis of HCC.
Keywords: hepatocellular carcinoma; liver cancer detection; Sigmis; FSV; machine learning.
Decision-making on the existence of soft exudates in diabetic retinopathy
by A. Reyana, V.T. Krishnaprasath, Sandeep Kautish, Ranjit Panigrahi, Mahaboob Shaik
Abstract: Medical image analysis is recognised to be the most important research area for diagnosis and screening of a wide range of medical problems. The commonly discussed diabetic retinopathy has become a vital factor of serious eye diseases leading to eye blindness. Diabetes mellitus is a metabolic disorder that is rapidly increasing worldwide. The major cause of Diabetic Retinopathy (DR) is the increase in blood insulin levels. The DR lesion thresholds inferred have protective effects and have no benefits for patients. The prediction of micro aneurysms from the fundus images is still the major challenge. The formation of micro aneurysms is the initial sign of DR to reduce the risk of non-proliferated DR. With this in mind, there exists a need for a diagnostic system for early detection of DR to be used by the ophthalmologist to identify different types of lesion, such as hemorrhages and exudates. This paper presents a new approach to detect and classify exudates in coloured retinal images, eliminating the replication of exudates by removing the optical disc region. Our research aimed to extend the current knowledge of several image processing techniques, including image enhancement, segmentation, classification, and registration for early diagnosis preventing visual impairment and blindness.
Keywords: diabetic retinopathy; homogeneity; segmentation; blindness.
Prediction of abnormal hepatic regions using ROI thresholding based segmentation and deep learning based classification
by Shubham Kamlesh Shah, Ruby Mishra, Bhabani Mishra, Om Pandey
Abstract: This paper proposes a novel Computer-Aided Diagnosis System (CADS) model using Artificial Intelligence (AI) to segment liver form abdomen CT scan. Deep Learning Convolutional Neural Network (DL-CNN) model is proposed to train the program to classify normal and abnormal liver images. For training dataset generation, another novel Region of Interest (ROI) based thresholding image processing technique is proposed. DL-CNN network is also compared with the basic CNN model to understand the difference between basic and deep learning networks. The basic CNN model yielded an accuracy of 50.00% while the DL-CNN model achieved an accuracy of 98.75%. It is also compared with other existing models, including AlexNet and adaBoostM1, and with classifiers such as na
Keywords: computer-aided diagnosis; DL-CNN; liver segmentation; image processing.
Orthogonal matching pursuit-based feature selection for motor-imagery EEG signal classification
by Rajdeep Chatterjee, Ankita Chatterjee
Abstract: This paper focuses on a framework that uses a small number of features to obtain high-quality classification accuracy of left/right-hand movement motor-imagery EEG signal. Motor-imagery EEG signal is filtered, and suitable features are extracted using a temporal sliding window-based approach. These features extracted from overlapping and non-overlapping approaches are further compared based on three different types of feature extraction technique: band power, wavelet energy entropy, and adaptive autoregressive model. The overlapping segments with wavelet energy entropy provide the best classification accuracy over other alternatives. The obtained classification accuracy is 91.43%, the highest ever reported accuracy for BCI Competition II dataset III. Subsequently, the orthogonal matching pursuit technique is used to select the subset of most discriminating features from the entire feature-set. It reduces the computation cost but still retains the quality of the classification results with only 1.43% information loss (that is, 90% classification accuracy), whereas the features-set size reduction is 75% for the same. It is found that the wavelet energy entropy technique performs consistently well in all the variants of our experiments and obtains a mean accuracy difference of 0.95% only.
Keywords: brain-computer interface; EEG; ensemble learning; orthogonal matching pursuit; motor imagery.
Remote homology detection using GA and NSGA-II on physicochemical properties
by Mukti Routray, Niranjan Kumar Ray
Abstract: Remote homology detection at amino acid level is a complex problem in the area of computational biology. Any pair of protein sequences are said to be homologous if they share common ancestry, have similar three-dimensional structures and exhibit similar functional similarity between proteins. These similarities can be detected using various laboratory techniques, such as X-ray crystallography, NMR spectroscopy etc. But these techniques are costly and time consuming. With the rapid growth in technological advancement, numerous sequences are generated day by day. Homology prediction of these sequences using laboratory techniques is becoming a tedious task. Hence there is a need to use machine learning algorithms to predict the homology of these unannotated protein sequences, which can save time and cost. This work is divided into three phases. Initially, the features are extracted from protein sequences using Principal Component Analysis (PCA) to build a chromosome set with representative features of each protein based on physicochemical properties. The second stage involves a genetic algorithm for the construction of a set of chromosomes for classification based on PCA and initialises the classifier to build up an error matrix. The third stage uses NSGA-II, crossover and mutation, and tournament selection for the next set of chromosomes. The output of this experiment is a set of minimum classification error values and minimum number of features used for classification of protein families. This technique is applied on the UniProt and SCOP 1.53 benchmark datasets. This approach tends to give superior accuracy over the profile based methods.
Keywords: principal component analysis; feature selection and classification; genetic algorithm; profile-based methods.
Special Issue on: Intelligent Healthcare Systems for Sustainable Development
Dermoscopic Image Segmentation Method Based on Convolutional Neural Networks
by Dang N. H. Thanh, Le Thi Thanh, Ugur Erkan, Aditya Khamparia, V B Surya Prasath
Abstract: In this paper, we present an efficient dermoscopic image segmentation method base on the linearization of Gamma-correction, and convolutional neural networks. Linearization of Gamma-correction is helpful to enhance low-intensity regions of skin lesion areas. Therefore, postprocessing tasks can work more effectively. The proposed convolutional neural network architecture for the segmentation method bases on the VGG-19 network. The acquired training results are convenient to apply the semantic segmentation method. Experimental results are conducted on the public ISIC-2017 dataset. To assess the quality of obtained segmentations, we make use of standard error metrics such as the Jaccard and Dice which are based on the overlap with ground truth, along with other measures such as the accuracy, sensitivity, and specificity. Moreover, we provide a comparison of our segmentation results with other similar methods. From experimental results, we infer that our method obtains excellent results in all the metrics and obtains competitive performance over other current and state of the art models for dermoscopic image segmentation.
Keywords: Dermoscopic Images, Deep CNNs, Machine Learning, Skin Lesions, Image Segmentation, Skin Cancer
Special Issue on: Advances in Computer Graphics and Imaging
Research on the Design of Visual Interface in Information Visualization
by Guangtao Ma, Tao Liu, Yang Zhou, Jun Li
Special Issue on: Machine Vision and Computational Intelligence in Recent Industrial Practice
Robust Skin Segmentation using Color Space Switching
by Ankit Chaudhary, Ankur Gupta
Special Issue on: Xxxx
A Query Driven Method of Mapping from Global Ontology to Local Ontology in Ontology-based Data Integration
by Haifei Zhang