An Integrated Deep Learning Framework for Tomato Leaf Disease Classification and Detect

Abstract

In this study, there is better DL and computer vision techniques or strategies to optimize tomato leaf ailment detection and growth agricultural productiveness.  The PCA DeepNet architecture represents a modern approach for shooting facts complexity and disease patterns with the aid of combining conventional ML with a customized DNN.  The research successfully categorizes wholesome and diseased tomato leaves utilizing the "faster region-based Convolutional Neural network (F-RCNN)," exhibiting excessive common precision and great classification accuracy.  The findings imply that the PCA DeepNet architecture surpasses opportunity approaches, doubtlessly providing good sized benefits to agricultural industries with the aid of accurately and reliably detecting ailments in tomato leaves.  This task complements its capability with the aid of incorporating elaborate models inclusive of DenseNet and Xception for classification, at the side of a voting Classifier, in addition to YoloV5 and YoloV8 for detection.  The goals are a type accuracy of 0.85 or more and a detection precision of 0.85 or more.

“Index terms – Machine learning, computer vision, convolutional neural networks, quicker region-based convolutional neural networks, generative adversarial networks, & diseases that affect tomato leaves”.