The paper presents a method that identifies and extracts features of chips on loading grinding wheel using image processing technology. The Sobel operator is adopted to detect chips edge. The segmenting threshold was obtained by applying Otsu's method. Image dilation connects the discontinuous segments to yield continuous and close shapes of chips. Erosion removes the noise from the image. The different optical characters between metal chips and abrasive grains are analysed. The ratios of chips are calculated and displayed to monitor the wheel surface working status. The toolbox of the MATLAB is used for image processing.