Presently, most target detection post-processing algorithms are short of self-adapting, which make them have many limitations in automatic target detection from the digital images. Therefore this paper provides a new type of automatic target detection post-processing algorithm (CASI) which adopts clustering algorithm based on swarm intelligence to post-process the detected target pixels. Because of the characteristics of self-organization, robustness, expandability and simplicity, CASI can self-adaptively restore the target pixels, divide them into different targets and delete the false alarm targets. Experimental results conform that the target detection post-processing effect of this new type of algorithm is much better than the ones of morphological transformation in common use.