A framework for Detection and classification of plant leaf Diseases using cnn optimized by evolutionary Methods

Abstract

The agricultural output would be a necessary element of the newlineeconomic development of any country; it provides raw resources, jobs, newlineand food for a large number of people. Crop yield variations are newlineexpected to change depending on a range of circumstances around the newlineworld. Among the most frequent causes is the use of chemical newlinefertilizers, but other factors including the existence of toxins in water newlinesources, variable precipitation regimes, and changes in soil quality. newlineAside from these problems, harvest problems have been among the most newlinecommon across the country, resulting in a substantial loss of yield. The newlinepresence of pests in plants reduces much of the production after newlinesupplying efficient materials to the fields. Therefore, accurate plant newlinedisease detection systems are becoming increasingly important. This newlineresearch proposes an innovative method for the prediction of plant leaf newlinediseases using images of different levels of diseases using Convolutional newlineNeural Network (CNN). newlineTo combat these ills, it is first essential to recognize and newlinefollow them on a large basis. Tensorflow with Keras and OpenCV has newlinebeen used to build a sensing approach to solve this problem. Because of newlinethe massive and scattered nature of computerized fields, classification newlinemust be executed and managed in a cloud computing environment. It newlinewas able to analyze plant images and detect common diseases. The newlineapproach made it possible to carry out a more precise verification using newlinethree convolution operations, ten terminals, and forty epochs. The newlinetraining data set included 11942 images, along with the audit setting, newlinevi newlineseveral 6421 images, or 35% of the length of the learning set. The use of newlinetechnologies to help stimulate agricultural activity and ensure food newlinesecurity. This solution uses OpenCV to identify image templates and newlineconvert these designs into information that Keras could use to build an newlineautomatic learning algorithm. Though there have been prior approaches newlineto solving similar difficulties, few designs that could specifically target.