A method for automatic segmentation of pork longissimus dorsi muscle (MLD) r egion from ribeye image was developed using KFCM clustering and improved water shed algorithm. Firstly, median filter and OTSU were used to remove noise and ba c kground. And then, kernel fuzzy Cmeans clustering (KFCM) was applied to remove fat pixels. Finally, hole filling operation and improved watershed algorithm wer e employed to segment the area of MLD. Sixty samples were used to test the perfo rmance of the proposed method. The success rate of segmentation was 8667%. Com p ared with the traditional morphology and watershed methods, it is proved that th e developed method could segment MLD region perfectly and avoid undersegmentat ion effectively. 