Image segmentation algorithm for disease detection of wheat leaves

Wheat diseases are harmful to wheat production, but there are few segmentation algorithms that can effectively identify common diseases of wheat leaves. This paper proposes an automatic and efficient solution with K-means clustering. Firstly, the colour image is transformed to Lab colour space from RGB. Clustering is then done by taking the absolute difference between each pixel and the clustering centre in Lab colour space. Unlike traditional methods, our method does not need manual setting for threshold value and is not affected by the selected channel. Our results show that the segmentation accuracy rates for three common diseases (powdery mildew, leaf rust and stripe rust) is more than 90%, which proves the efficacy of our method.