Rice Disease Identification and Classification by Integrating Support Vector Machine With Deep Convolutional Neural Network

Over the world, rice is one of the most extensively consuming staple foods. But due to various diseases of rice, farmers can't yield rice according to their expectation. Therefore, the quality, quantity and production of rice are being interrupted. It is difficult for the farmers to diagnose diseases, because of the lack of knowledge and unavailability of professional specialists. It also requires much time in manually identifying and classifying rice diseases. Therefore, an automated and accurate identification system has become essential to overcome this problem. A unique and fruitful approach has been presented to detect rice diseases through the application of an Artificial Intelligence (AI) and effective computer vision techniques. In this paper, we have adopted the SVM classifier in order to combine with an AI model named deep CNN. The transfer learning technique has been used to improve the proposed model. After that 1080 images of our dataset of nine different rice diseases are used to re-train our proposed model. SVM classifier is then trained with the features which are extracted from the DCNN model. The proposed model proficiently identified and classified rice diseases of nine different types and achieved 97.5% accuracy.