Detection of Plant Disease using ResNet framework in Comparison with Neural Network Classifier to Improve Classification Accuracy

In order to reach the objective of enhancing the accuracy of plant disease diagnosis, the suggested study seeks to make use of ResNet as opposed to a New Neural Network classifier. This will be done in order to achieve the goal. In the process of identifying plant diseases, the algorithms ResNet (N=10) and New Neural Network classifier (N=10) are used as diagnostic tools. The sample size for each sample is determined with the use of a G power calculator, and the sample size that is taken into consideration for each sample is 10. The findings acquired by the New Neural Network classifier were only 90% accurate, while the ResNet approach displayed significantly greater performance, with an accuracy of 95%. This was in comparison to the results obtained by the New Neural Network classifier. A significant accuracy value of p=0.001 (p0.05) was discovered to be acquired via the use of SPSS Statistical Analysis. This was proven to be the case. As compared to the New Neural Network classifier, the ResNet classifier performs much better when used to the categorization of plant diseases.