An Evaluation Method of Dynamic Camouflage Effect Based on Multifeature Constraints

Abstract

Most of the existing camouflage effect evaluation methods are for static images, and the evaluation methods have problems of singularity and subjectivity. Therefore, this paper takes the camouflage of moving objects in video as the research object and proposes a comprehensive camouflage effect evaluation method based on multifeature constraints. This method has two parts: the Homo-F (homography transformation and optical flow) target detection module and the camouflage effect evaluation module. The former uses the optical flow method to correct the target detection results obtained by the homography transformation. The latter performs statistical analysis on the target detection results and the feature information of the neighborhood background and describes the effect of camouflage from multiple angles such as the degree of target fusion, the repetition rate and the target detection stability in the video sequence. The experimental results of the comprehensive camouflage evaluation of moving targets show that the proposed method can objectively and accurately evaluate moving targets with different levels of camouflage, which verifies the reliability and effectiveness of the method.