Published Paper Details:

For detecting the incidence and quantifying the severity of variability in crops, image-based plant disease detection is one of the most important activities in precision agriculture. Pathogen-caused diseases account for 70 percent to 80 percent of the variabilities, with 60 percent to 70 percent appearing on the leaves compared to the stem and fruits. This study uses leaf image data to conduct a comparative comparison of model implementation for plant disease diagnosis. Until recently, most of these image processing algorithms relied on what some referred to as "shallow" machine learning architectures, and some still do. The DL network is quickly becoming the gold standard for image recognition and pattern analysis research. Regardless, there aren't many research on its use in detecting disease in plant leaves. Thus, we used conventional settings and took into account the three critical elements of architecture, processing capacity, and the amount of training data in our study on a large plant leaf disease picture dataset. In a developing agricultural economy like India, early detection of plant leaf detection is critical. Leaf diseases in plants must be diagnosed at an early stage and predictive techniques implemented to make them safe and avoid losses to the agri-based economy, not only as an agricultural economy but also as one with a big population to feed. This work proposes to identify Leaf disease utilising image processing techniques based on image segmentation, clustering, and image detection algorithms, all of which contribute to a reliable, safe, and accurate leaf disease detection system with specialisation.