Abstract: For micro-gap weld joint whose width is less than 0.1 mm, the weldments are magnetized by using an excitation magnetic field. Meanwhile, a magneto optical sensor based on the principle of Faraday magneto effect is applied to acquire the magneto optical images of weld joint. An edge detection method which fuses the wavelet multi-scale information of image is developed to deal with the magneto optical images of weld joint. Three layers wavelet decomposition of the weld magneto optical image is used to obtain wavelet image which contains high frequency information of weld edge. And then it fuses all high frequency images according to the richness of details which contain the high frequency information of each scale. By using wavelet modulus local maxima to detect the edge of fused image, the smooth weld edge with good noise immunity and continuity is obtained. The result of seam tracking test shows that the proposed method is an effective way to extract weld edge and is suitable for the image processing in micro-gap seam tracking based on magneto-optical imaging sensing.