Cutting-edge AI for Plant Health: Advanced Residual Neural Networks in Leaf Diseases Segmentation and Classification

Agricultural Farming is the backbone of India. Farmers face many challenges in agricultural farming to get high yield of production. In agriculture, plant diseases lower mass productivity. Farmers always experience many challenges in satisfying their requirements as a result of major losses in agriculture. The main objective of this study was the plants grown in the Kanyakumari region in order to evaluate the pH of the soil. This paper analyses various diseases that impact on the categories of plants like creeper, climber and shrubs based on Monocot and Dicot family. The existing system focused on the various classification algorithms with various parameters but with the drawback of less accuracy and high computational time. Despite this, a variety of techniques have to be used in order to detect plant illnesses early on and improve productivity in agriculture. More algorithms using machine learning are at present readily available and all of them have the potential to be improved to provide better accuracy levels. For the purpose of segmenting and classifying the affected leaves, the hybrid model is introduced that combines Deep Convolutional Neural Networks (DCNNs) with ResNets. A comparison with existing models has been conducted and the proposed method offers a higher accuracy rate of 99.72%.