The establishment of fast and efficient blasting fragmentation detection technology and identification method can quickly quantify and feedback the blasting effect. Aiming at the false edges or missing edges in the edge detection operator of blasted ore images, an improved Canny operator edge detection method was proposed in this paper.Using images of blast piles from Yuanbaoshan open pit mine as the data source, firstly, statistical filtering was used to improve the image edge blur problem that may be caused by Gaussian filtering when denoising; secondly, the gradient amplitude and gradient direction of the burst image were calculated based on the 3×3 gradient template of the operator; finally, based on the natural iteration method, the image threshold was automatically obtained, and the discontinuous edge was repaired by the morphological closing operation and thinning operator to obtain the final detection edge.The results demonstrate that this method achieves precise image segmentation, enhancing the accuracy of large block rate detection in blasting operations.