Identify and classify pests in the agricultural sector using metaheuristics deep learning approach
Atul B. Kathole, Jayashree Katti, Savita Lonare, Gulbakshee Dharmale
Abstract
The most significant challenges impacting agricultural production and preventing the sustainable expansion of the agricultural sector are insect pests and crop diseases. It’s inefficient to put surveillance cameras in areas densely populated by the pests you’re trying to catch, and it’s typically not enough to check in on the photos generated by your Internet of Things monitoring devices from a single location. Although the Internet of Things (IoT) is a specialised technology and analytics system, it has found many applications, including in agricultural contexts. To aid in identifying and naming the insects seen in the pictures, this study seeks to establish a model for pest identification and categorization. In the beginning, data is gathered by Internet of Things devices using the IoT platform. The collected images are then subjected to an object detection analysis using Yolov3. The Adaptive Honey Badger Algorithm is used to fine-tune the classifier’s parameters. (AHBA). These findings demonstrate that the proposed technique is preferable due to its ability to speed up the collection of agricultural data and guarantee technical support for population estimation and pest monitoring.