基于背景还原的红外伪装效果评价方法  

Evaluation method of infrared camouflage effect based on background restoration

作　　者：赵晓枫 吴飞 徐叶斌 牛家辉 蔡伟 张志利 ZHAO Xiaofeng;WU Fei;XU Yebin;NIU Jiahui;CAI Wei;ZHANG Zhili(Missile Engineering College,Rocket Force Engineering University,Xi’an 710025,China;Armament Launch Theory and Technology Key Discipline Laboratory of China,Xi’an 710025,China)

机构地区：[1]火箭军工程大学导弹工程学院,陕西西安710025 [2]兵器发射理论与技术国家重点学科实验室,陕西西安710025

出　　处：《系统工程与电子技术》2022年第8期2554-2561,共8页Systems Engineering and Electronics

基　　金：国家自然科学基金(41404022)资助课题。

摘　　要：针对复杂环境背景下目标红外伪装效果评价问题,提出了一种基于背景还原的红外伪装效果评价方法。首先利用改进的Criminisi算法对目标遮蔽下的背景区域进行修复,得到背景还原图像,然后与原图进行相似性度量;考虑到实际还原背景与理想还原背景之间因图像修复算法局限性而引起的偏差,使用IL-NIQE模型对背景还原图像的修复质量和还原效果进行评价;最后将相似性度量值与还原效果度量值进行非线性融合,完成目标红外伪装效果的综合评价。实验表明,该评价方法符合人眼视觉特性,能够对复杂环境背景下目标的红外伪装效果进行较好的客观评价。Aiming at the problem of infrared camouflage effect evaluation of targets in complex environment, an infrared camouflage effect evaluation method based on background restoration is proposed. The improved Criminisi algorithm is used to repair the background area under the target shielding, and the background restored image is obtained. Then the similarity between the restored background and the original infrared camouflage target is measured. Considering the deviation between the restored background and the real masking background, use the IL-NIQE model to pre-evaluate the restoration quality and restoration effect of the background restored image, and obtain the restoration effect measurement value. A nonlinear fusion method is adopted to fuse the restoration effect measure and the similarity measure to complete the comprehensive evaluation of the target infrared camouflage effect. Experiments show that the evaluation method accords with human visual characteristics and can objectively evaluate the infrared camouflage effect of targets in complex environment.

关 键 词：图像处理 红外伪装评价 图像修复技术 非线性特征融合 

分 类 号：E951.4[军事—军事工程]