Agriculture is a key source of livelihood. However, the agricultural sector is declining across the country, which has an adverse effect on the quantity of food products that can be harvested. In developing nations like India, agriculture offers an extensive range of job opportunities for the villagers. India's agriculture is composed of many crops, and according to a survey, nearly 70% of the population depends on agriculture. Early detection of diseases is crucial. Faster and more precise prediction of leaf diseases could help reduce losses. The opportunity to enhance the accuracy of object detection and recognition systems has been made possible by the tremendous advancements and discoveries in deep learning. To address this problem, a Deep Convolutional Neural Network (CNN) architecture has been developed to accurately identify pathogens in plant leaves. The study focuses on five major crops in the South Asian Agriculture Region, namely tomato, potato, pepper, okra, and wheat, using leaf pathogen images from the Plant Village Dataset. A total of 17,430 images were collected for a proposed methodology. In this study, the images were segmented into four different training samples, underwent a pre-processing stage, and finally, the leaf diseases were identified using a proposed method. The obtained accuracy for 5% of a training sample of tomato, potato, pepper, okra, and wheat is 89.6%, 93.2%, 80%, 76%, and 88%, respectively.

Deep Learning, CNN, Leaf Disease Detection, Tomato, Potato, Okra, Pepper and Wheat