Abstract:

Plant diseases cause significant crop losses globally, posing challenges to agricultural productivity. Detecting these diseases is difficult due to the lack of expert knowledge. Deep learning-based models offer promising solutions using leaf images, but issues like the need for larger training sets and computational complexity persist. To address this, we propose a convolutional neural network (CNN) with fewer layers, reducing computational burden. Augmentation techniques such as shift, shear, scaling, zoom, and flipping are applied to expand the training set without capturing more images. As agriculture remains crucial for nourishing about half of the global population, increasing production by 50-60% is urgent, especially in regions with rapid population growth. Despite an expanding cultivation area, apple crop production in India faces challenges, with minimal growth in yield. In Himachal Pradesh, a major apple-producing state, fungal diseases significantly impact fruit quality. Our project addresses these challenges by employing deep learning models, including pre-trained ones, and utilizing YOLO series models for efficient disease detection in apples. By leveraging image processing and AI, timely and accurate disease diagnosis is ensured. This project has the potential to revolutionize disease detection in apple plants, enhancing food security globally. Farmers stand to benefit from prompt intervention, safeguarding their crops and ensuring increased yields, thereby contributing to overall food security for the growing global population.

Description:

Apple diseases, classication, convolutional neural network, deep learning, disease detection, image processing, machine learning.