Parallel Deep Convolution Neural Network (P-DCNN) Prediction of Paddy Crop Disease

Agriculture is one of the major sources for gross regional product among the developing nations. The nature and amount of crop cultivated along with its yield in each region varies with soil and climate. Additionally, the crop yield is affected due to the event of disease. Paddy is one of the chief food crops cultivated in numerous parts, and it succumbs to several diseases. The framers often find it hard to predict the nature of disease over the paddy crop. For supporting the farmer, a novel Parallel Deep Convolution Aggregation Neural Network (P-DCNN) classifier is proposed to predict and localize the disease in paddy crop using raw images. The novel parallel classifier which has hybrid tanh and sigmoid function is developed for the proposed work. The input images considered for processing are given into equally divided and sent to two smaller networks works independent to each other simultaneously, and then outcome is merged into a final output layer. Parallelization reduces the computational time for processing the images through the network. The performance of the proposed framework is evaluated on various performance metrics.