The invention discloses a partitioning self-adaptation image segmentation method based on FCM. The partitioning self-adaptation image segmentation method based on the FCM specifically comprises the steps that an image to be segmented is input; segmentation is carried out on the image to be segmented by using an image segmentation method based on the standard FCM, partitioning processing is carried out on a result image, numbering is carried out on image region blocks, and variances of the image region blocks are calculated; a neighborhood averaging method is used for processing the image to be segmented, then, segmentation is carried out by using the image segmentation method based on the FCM, partitioning processing is carried out on a result image, numbering is carried out on image region blocks, and variances of the image region blocks are calculated; the variances of image blocks of one segmented image and the variances of image blocks, at the positions corresponding to the image blocks of the segmented image, of the other segmented image are compared, and the image blocks with the smaller variances are selected as the final segmentation result. According to the partitioning self-adaptation image segmentation method based on FCM, gray information and space information of image pixels are taken into full consideration, segmentation methods are selected for the image to be segmented in a partitioning self-adaptation mode, segmentation quality of the image with noise is improved, and the image segmentation result which is more accurate is obtained.