Abstract: Digital image processing technology was used to measure particle size of uranium ore fragmentation and determine particle size distribution. Image guided filter was applied to ore image filtering for the first time, which could filter image noise and preserve ore image edge information. PCNN image segmentation algorithm based on between-class posterior maximum cross entropy criterion was used to segment ore image,which could reduce ore adhesion phenomenon. In order to solve ore adhesion phenomenon after first image segmentation, a digital image cutting algorithm based on concave point matching was proposed to secondly segment adhesive ore image, which could effectively separate adhesion ore image.The parameter measurement method based on shape features was used to measure particle parameters, which could improve measurable accuracy of particle parameters. The statistical distribution charts of ore particle size were drawn with particle parameters. Experimental data show that the method can improve measure accuracy and satisfy with practical needs.