The longitudinal tearing of the conveyor belt will cause huge economic losses, so the real-time and accuracy of detecting the tearing of the conveyor belt is particularly important during the production of coal. Based on the advantages of OpenCV open source platform and function library, a wavelet fusion method for conveyor edge detection is proposed. Wavelet multi-scale decomposition is used to decompose the image into low-frequency region and high-frequency band. The low-frequency region of multi-scale multi-directional Susan morphological filtering and the high frequency band with improved Canny operator are respectively extracted to the information of crack texture. Finally, image fusion is performed. The technique, inverse wavelet reconstruction original, reconstructs the new original image. In addition, the image is evaluated and analyzed by parameters such as root mean square error and peak signal-to-noise ratio. The experimental results show that the information of anti-noise and crack extraction is better than the traditional detection method.