Title: Big data analytics in agriculture: cloud-based architecture for crop disease classification

Authors: P. Geetha; S. Thaiyalnayaki; S.J. Vivekanandan; G. Abirami

Addresses: Department of Computational Intelligence, School of Computing, SRM Institute of Science and Technology, KTR Campus, Chennai, India ' School of Computing, Bharath Institute of Higher Education and Research, Chennai, 600 073, Tamil Nadu, India ' Department of Computer Science and Engineering, Dhanalakshmi College of Engineering, Sriperumbudur, 601301, Tamil Nadu, India ' Department of Computer Science and Engineering, AMET Deemed to Be University, Chennai, Tamil Nadu, India

Abstract: This research developed a hybrid methodology for crop disease classification using cloud computing and big data analytics in agriculture. First, adaptive bilateral filtering is suggested during the preprocessing stage for acquired data from the plant village dataset. After that, the MapReduce framework for feature extraction like improved text on, GLCM, and LGXP-based features are extracted under the mapper phase. In the reduction phase, these features from each mapper are combined and delivered to the classification process. At last, the classification process is carried out by the hybrid model with the combination of Improved deepmaxout and DCNN classifiers, named as N-Sigmoid activated Maxout-Convolutional network (N-SAMCN) is used to classify the image. Moreover, the experimental analysis of the proposed hybrid model on the image classification process over the traditional classifiers.

Keywords: bilateral filtering; GLCM; LGXP; improved texton features; modified k-means clustering; crop disease classification; deepmaxout.

DOI: 10.1504/IJAHUC.2024.142705

International Journal of Ad Hoc and Ubiquitous Computing, 2024 Vol.47 No.4, pp.191 - 208

Received: 16 Sep 2023
Accepted: 20 Feb 2024
Published online: 18 Nov 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article