A Novel Approach For Silkworm Disease Detection And Classification By Using CNN And Image Processing Techniques

Abstract

Silkworms, scientifically known as Bombyx mori, are integral to the silk industry, which has substantial economic and cultural significance in various regions worldwide. However, silkworms are susceptible to several diseases that can severely impact their health and productivity, leading to significant economic losses. Common silkworm diseases include Flacherie, Grasserie & Muscardine, each caused by different pathogens such as bacteria, viruses, and fungi. Effective identification & timely behaviour of these ailments are crucial to maintaining the health of silkworm populations and ensuring consistent silk production. For this we propose an automated system for the identification of silkworm diseases and providing treatment recommendations using CNN and image processing. Leveraging deep learning techniques, the system is considered to accurately classify images of silkworms into categories of health status, including specific diseases such as Flacherie, Grasserie, Muscardine, and a healthy state. Experimental results analysis demonstrates 99% of accuracy using proposed model. This approach signifies a step forward in agricultural technology, demonstrating the potential of AI in solving practical problems in sericulture.