Crop disease monitoring and recognizing system by soft computing and image processing models

In order to make the crop disease intelligent diagnosis system more cheap, convenient and efficient for common farmers, an effective and general monitoring system of crop diseases is constructed by Internet of Things (IoT). In the system, crop disease images are collected by IoT and passed to the Web server by wireless network. First, a crop disease image dataset is constructed. Second, a K-mean clustering algorithm is utilized to segment disease leaf images, and the sum and difference histogram (SADH) feature vector is extracted from each segmented defect image based on the intensity values of the neighboring pixels. Finally, a reasoning process is built on the reasoning decision tree for monitoring system. The results validates that the proposed system is benefit for monitoring and controlling crop diseases in practice.