Controlling of Pest using Image Processing and Machine Learning Methods

In addition to giving humans food, agriculture is a major source of resources for every nation’s economy. Pests and insects cause damage to crops, making them extremely risky for the crop’s overall growth. In the world of agriculture, early tormentor identification may provide significant challenges. Using insecticides is the greatest way to control the tormentor infection. On the other hand, overuse of pesticides is detrimental to animals, plants, and even the general population’s automatic method of early insect detection. Digital image processing techniques are widely used in agricultural research and offer great perspectives, especially in the field of plant protection, which leads to crop management. This study discusses a novel approach to pest monitoring through early detection. Using a photography camera, one may create nonheritable images of pest-ridden leaves. An automated system that can identify the type of pests on crops in addition to examining the crops to detect infestations of tormentors is required. Support Vector Machine (SVM) is utilized for the categorization of images with and without pests, while the YOLO algorithmic method is used for tormentor detection.