EARLY PEST DETECTION FROM CROP USING IMAGE PROCESSING AND COMPUTATIONAL INTELLIGENCE

Abstract

In the agricultural sector, one of the significant challenges revolves around promptly 
identifying and addressing pest infestations. While insecticides offer a convenient 
solution, excessive use poses risks to both human health and the ecosystem. Integrated 
pest management aims to prevent parasite infections through a combination of 
physical and biological methods. Within agricultural research, digital image 
processing plays a crucial role, especially in ensuring plant security and productivity. 
This study explores a new approach for early parasite detection, utilizing digital video 
cameras to capture images of affected leaves. Through attribute extraction and picture 
classification algorithms, insect presence on leaves is detected and refined to 
grayscale images. These images are then processed using MATLAB software, 
employing feature extraction techniques. Support Vector Machine classifiers are 
utilized to identify the types of pests present, aiding in targeted pest control strategies.