In order to detect the shape and position information of detonation transmission tubing more accurately, edge detection is an essential step. Aiming at the problem that the traditional Canny algorithm is not effective in preserving the edge of the target and the extraction of the edge of the detonator is not accurate, an improved method is proposed. Firstly, the improved edge sensing weight is used to guide the filter to remove the image noise. Then, the gradient of the filtered image is combined with the adaptive block variance weighting to extract the feature of the image target. Finally, the Gaussian function algorithm of particle swarm optimization is used to simulate the gray distribution of the image to obtain the threshold adaptively, and the edge of the image is extracted by breaking the artificial limitation. Experimental data show that compared with the original filtering algorithm, the improved edge sensing weight guided filter improves the PSNR by 5-16dB. Under the same conditions, compared with t