Addressing the prevalent issues of subjectivity and the limited scope of criteria in traditional camouflage assessment,this paper in-troduces an integrated evaluation method for facial camouflage that combines visual features with Deep Neural Networks(DNN).This method integrates edge detection,similarity metrics,and DNN-based facial recognition to establish a comprehensive evaluation system.Validated through simulation experiments in four natural environments,this study effectively quantifies the effectiveness of facial camouflage in various backgrounds.It offers valuable guidance for future design and implementation of facial camouflage,positively influencing its practicality and effectiveness.
维普
万方数据