The detection of camouflage color moving objects in complex scenes is currently a hot and crucial topic, which has significant implications in the military field and other special scenarios. A method for camouflage color moving object detection based on corner detection and optical flow field was proposed. Firstly, grayscale conversion, denoising, and edge detection were undergone for the image. Secondly, corner detection and pyramid Lucas Kanade optical flow method were used to detect camouflage color moving objects. Thirdly, a combination of Canopy clustering algorithm and K-means clustering algorithm was used to cluster the detected optical flow points. Finally, considering the high number and dense distribution of optical flow points of camouflage color moving objects, K classes obtained through clustering algorithm were screened to ultimately determine the camouflage color moving objects. The experimental results showed that the method can effectively detect camouflage color moving objects.