Article: A behaviour detection algorithm integrating lightweight networks and feature recombination Journal: International Journal of Sensor Networks (IJSNET) 2026 Vol.50 No.2 pp.118 - 132 Abstract: Traditional behaviour detection methods often have problems such as low accuracy and slow processing speed, making it difficult to meet the practical application needs of industrial production scenarios. This study proposes a behaviour detection algorithm that integrates lightweight networks and feature recombination. First, we replace you only look once (YOLO) backbone with an enhanced MobileNetV3, reducing model complexity and accelerating inference. Second, we introduce content-aware reassembly of features, replacing conventional upsampling to improve precision. Further, switchable atrous convolution in the neck network enhances adaptability to multi-scale features, while vision transformer with deformable attention strengthens spatial modelling. Ablation experiments demonstrate the algorithm's effectiveness, achieving a 75.2% mAP, with gains of 2.8% and 4.8% in precision and recall, respectively. Compared to existing technologies, this method offers the advantages of fast speed and high accuracy, making it suitable for real-time detection scenarios, such as those in the petrochemical industry. Inderscience Publishers - linking academia, business and industry through research

Title: A behaviour detection algorithm integrating lightweight networks and feature recombination

Authors: Gen Liang; Yu Zhang; Guoxi Sun; Xinchao Li

Addresses: Guangdong Provincial Key Laboratory of Petrochemical Equipment Fault Diagnosis, School of Science, Guangdong University of Petrochemical Technology, Guangdong, 525000, China ' School of Electronic Information Engineering, Guangdong University of Petrochemical Technology, Guangdong, 525000, China ' School of Electronic Information Engineering, Guangdong University of Petrochemical Technology, Guangdong, 525000, China ' School of Electronic Information Engineering, Guangdong University of Petrochemical Technology, Guangdong, 525000, China

Abstract: Traditional behaviour detection methods often have problems such as low accuracy and slow processing speed, making it difficult to meet the practical application needs of industrial production scenarios. This study proposes a behaviour detection algorithm that integrates lightweight networks and feature recombination. First, we replace you only look once (YOLO) backbone with an enhanced MobileNetV3, reducing model complexity and accelerating inference. Second, we introduce content-aware reassembly of features, replacing conventional upsampling to improve precision. Further, switchable atrous convolution in the neck network enhances adaptability to multi-scale features, while vision transformer with deformable attention strengthens spatial modelling. Ablation experiments demonstrate the algorithm's effectiveness, achieving a 75.2% mAP, with gains of 2.8% and 4.8% in precision and recall, respectively. Compared to existing technologies, this method offers the advantages of fast speed and high accuracy, making it suitable for real-time detection scenarios, such as those in the petrochemical industry.

Keywords: behaviour detection; lightweight network; feature reorganisation; dilated convolution; deformable attention; DAttention.

DOI: 10.1504/IJSNET.2026.152042

International Journal of Sensor Networks, 2026 Vol.50 No.2, pp.118 - 132

Received: 18 Jul 2025
Accepted: 06 Aug 2025
Published online: 04 Mar 2026 *

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

Title: A behaviour detection algorithm integrating lightweight networks and feature recombination

Keep up-to-date