Agricultural monitoring system in video surveillance object detection using feature extraction and classification by deep learning techniques. (September 2022)

• Record Type: Journal Article
• Authors: Khan, Shakir; AlSuwaidan, Lulwah
• Is Part Of: Computers & electrical engineering. Volume 102(2022)
• Publication Date: 2022-09
• Subjects: Video object detection -- Human action recognition -- DL -- Surveillance camera -- Classification; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2022.108201
• Languages: English
• ISSNs: 0045-7906

Abstract

Highlights: Systems that can infer crop state from low-cost sensing devices are required for agricultural applications such as yield prediction, precision agriculture, and autonomous harvesting. This research is based on agriculture monitoring system of real-time video frames processing. The aim is to collect the agriculture surveillance video based on IoT module and create the dataset by processing them into video frames as input. Finally, to evaluate the various dataset in terms of accuracy, precision, recall, F-1 score, RMSE. Abstract: Systems that can infer crop state from low-cost sensing devices are required for agricultural applications such as yield prediction, precision agriculture, and autonomous harvesting. This research is focusing on agriculture monitoring system of real-time video frame processing, extraction and classification with deep learning techniques. Here the input video data has been collected from agriculture surveillance camera and transmitted through IoT module. Then the feature was extracted using probability-based lasso network regression (Pr_La-Net_Reg). Then the extracted feature has been classified using Dynamic radial functional neural network (Dy_Rad_FuNN) based Deep learning architecture. For MATLAB R2018a, an HP envy machine with Intel core i5, RAM 4 GB DDR2-RAM, Digital Camera 16 Mega Pixel, and disk space 8 GB was used to validate the system. Three benchmark datasets for video frame- based evaluation are used to conduct the suggested approach's qualitative and quantitative experimental studies. Graphical abstract: Image, graphical abstract