摘要
迷彩伪装效果评价的多指标权重分配多为平均分配或专家评判,难以客观、科学地评价伪装效果。提出了基于模糊聚类方法、结合信息熵和统计特征的多指标权重分配算法。该算法通过模糊聚类得到样本的分类结果,应用信息熵和统计理论计算各指标的权重分配。对以结构、纹理、颜色、二阶矩特征等为指标的迷彩伪装样本进行了权重分配实验,分析比较了专家评判法、熵权法和该算法确定的权重之间的差异,并通过实验验证了算法的有效性和科学性。
The multi index weight allocation of camouflage effect evaluation is usually allocated equally or given directly by experts, which has strong subjectivity and is difficult to evaluate the camouflage effect objectively and scientifically. In order to deal with this problem, we propose a multi index weight allocation algorithm based on fuzzy clustering method, combined with information entropy and statistical features. The method obtains the classification results of the samples by fuzzy clustering, and calculates the weight allocation of each index by using information entropy theory and statistical theory. We carry out the weight allocation experiment on camouflage samples with structural similarity, texture similarity, color similarity and second moment matrix. Finally we analyze and compare the differences of the weight determined respectively by expert evaluation method, entropy weight method and the algorithm proposed in this paper. We design a new weight allocation algorithm, and verify the effectiveness and scientificity of the algorithm by experiment.
作者
李中华
喻钧
胡志毅
代军
张文琪
LI Zhong-hua;YU Jun;HU Zhi-yi;DAI Jun;ZHANG Wen-qi(School of Computer Science and Technology,Xi’an Technological University,Xi’an 710021,China;Institute of Engineering Design and Research Institute of the Army Research Institute,Beijing 100043,China)
出处
《火力与指挥控制》
CSCD
北大核心
2019年第10期49-54,共6页
Fire Control & Command Control
基金
陕西省工业科技攻关基金资助项目(2015GY059)
关键词
模糊聚类
信息熵
权重分配
迷彩伪装评价
fuzzy clustering
information entropy
weight allocation
camouflage evaluation
