面向视觉相似的红外伪装仿真效果评估

针对红外伪装仿真系统效果评估困难的问题,提出了一种面向视觉相似的解决方法。从视觉相似的角度出发,定义了一种新的视觉相似性模型,并建立了包括温度偏差系数、离散偏差系数和视觉相似性系数的评价指标体系;为减小主观经验的影响,引入直觉模糊集理论,建立了仿真效果评估的多属性决策模型;采用逼近理想解的排序算法(TOPSIS)进行模型求解,并针对算法存在距离正、负理想方案区分度不高的问题,定义了一种基于直觉模糊集Hamming距离的灰色接近系数来表征方案与理想方案的接近程度,有效解决了多属性决策问题。实例表明,该方法的评估结果与TOPSIS方法的结果一致,且与专家评估结果一致,验证了评价指标体系的合理性和评估方法的可行性。

The Effect Evaluation of Infrared Camouflage Simulation System Based on Visual Similarity

An evaluation method based on visual similarity is presented to solve the difficulty in the effect evaluation of infrared camouflage simulation system.From the perspective of visual similarity,a new visual similarity model is proposed,and the evaluation index system,including temperature deviation coefficient,discrete deviation coefficient and visual similarity coefficient,is established.A multi-attribute decision making (MADM) model based on intuitionistic fuzzy sets for evaluation is built to reduce the influence of subjective experience.The technique for order preference by similarity to an ideal solution (TOPSIS) is used to solve the model.For the same distance to idea/anti-idea ahernative,the grey approach coefficient based on intuitionistic fuzzy sets and the Hamming distance is used to describe the similarity between the alternatives,solving the MADM problem effectively.The application example shows that the evaluated results of the proposed model are consistent with the evaluated results of TOPSIS method and experts,thus verifying the reasonability of evaluation index system and the feasibility of the proposed method.