Plant Disease Detection and Classification Using Hybrid Model Based on Convolutional Auto Encoder and Convolutional Neural Network

During its growth stage, the plant is exposed to various diseases. Detection and early detection of crop diseases is a major challenge in the horticulture industry. Crop infections can harm total crop yield and reduce farmers’ income if not identified early. Today’s approved method involves a professional plant pathologist to diagnose the disease by visual inspection of the afflicted plant leaves. This is an excellent use case for Community Assessment and Treatment Services (CATS) due to the lengthy manual disease diagnosis process and the accuracy of identification is directly proportional to the skills of pathologists. An alternative to conventional Machine Learning (ML) methods, which require manual identification of parameters for exact results, is to develop a prototype that can be classified without pre-processing. To automatically diagnose tomato leaf disease, this research proposes a hybrid model using the Convolutional Auto-Encoders (CAE) network and the CNN-based deep learning architecture of DenseNet. To date, none of the modern systems described in this paper have a combined model based on DenseNet, CAE, and Convolutional Neural Network (CNN) to diagnose the ailments of tomato leaves automatically. The models were trained on a dataset obtained from the Plant Village repository. The dataset consisted of 9920 tomato leaves, and the model-to-model accuracy ratio was 98.35%. Unlike other approaches discussed in this paper, this hybrid strategy requires fewer training components. Therefore, the training time to classify plant diseases with the trained algorithm, as well as the training time to automatically detect the ailments of tomato leaves, is significantly reduced.