In order to improve the segmentation accuracy and reduce the segmentation running time of Gaussian mixture model used on wheat lesion images, a segmentation method based on PCA and Gaussian mixture model was proposed. Firstly, in order to completely use the color information of an image, three primary color channels of the image were obtained through the principal component analysis (PCA) method from R, G, B orH, S, Vcolor channels of this image. Secondly, the image was divided into many blocks, which were then sorted according to their mean pixel values. After sorting, those blocks lying in the front and the rear were selected to comprise a new pixel set by the Gaussian mixture model, and further, the corresponding Gaussian model parameters were obtained. Finally, the proposed method traveled all pixels in the image and classified each pixel into the corresponding Gaussian model category. Experimental results show that the proposed method has gained better promotions in segmentation error rate and running time compared with the traditional segmentation method and is effective for wheat leaf rust lesion segmentation.