An efficient plant disease prediction model based on machine learning and deep learning classifiers

Agriculture is the majorsource of the nation's economic growth, but the emergence of certain plant-related diseases has a negative effect on the sector's production. Diagnosing plant diseases is crucial to solving this issue, teaching farmers how to prevent diseases, and implementing effective management. The total yield may be negatively impacted by diseases if they are not discovered early, which would reduce the farmer's profits. Numerous researchers have presented various cutting-edge systems based on numerous techniques to address this issue. Numerous approaches have been used already by researchers for this purpose, but some techniques relating to vision have not yet been investigated. Six cutting-edge models for plant disease detection based on machine learning and deep learning were briefly deliberated in this research to analyze the efficiency of each approach. In this research, the Resnet-101 model is used to efficiently extract the feature, and the accuracy, sensitivity, and specificity of the metrics are measured to demonstrate the efficiency of each model. Recent research has shown average accuracies of 94.38%, 94.785, and 92.45% in the detection of Apple, potato, and strawberry diseases respectively.