Aiming at the traditional optical camouflage effect assessment, it is more difficult to objectively reflect the camouflage effect of camouflage target in a specific background, and it is easy to ignore the bad camouflage in a specific angle, etc., it puts forward a multi-angle optical camouflage assessment method based on the saliency of the image. The brightness, color and frequency features of the image are extracted to obtain each feature saliency map, the weight of each saliency map is calculated using the entropy weight method, and the multi-feature fusion saliency map is integrated to obtain the multi-feature fusion saliency map, and the saliency coefficients of each angle are calculated by analyzing the performance of the target area in the multi-feature fusion saliency map to comprehensively assess the camouflage effect of this camouflage scene and to experimentally validate it by using the optical images captured in two typical backgrounds. Experimental validation. The validation results show that the method is correct and effective, and the camouflage scheme can be optimized according to the evaluation results of specific angles.