A detection of tomato plant diseases using deep learning MNDLNN classifier

In the world, tomato is a significant economic crop. However, it is easily affected by various diseases. Misprediction of disease is caused since many prevailing methodologies focused on the tomato plant's specific portion. Thus, by employing deep learning (DL) multivariate normal DL neural network (MNDLNN) classifier, the study has proposed a framework for tomato plant disease (PD) detection. Firstly, the input images' colours are transmitted into HSI format. Next, from the images, the green pixels are masked, and healthy and unhealthy regions are isolated. Next by deploying the region of interest (ROI), the fruit and root are detected. Then, by utilizing the rectilinear K-means (KM) clustering (RKMC) algorithm, the unhealthy regions are segmented. Afterwards, by utilizing random motion squirrel search optimization (RMSSO), the essential features are extracted. Finally, MNDLNN effectively detects and classifies the disease types. The results revealed that the proposed framework performed the disease detection process more precisely than other top-notch methodologies.