基于超效率数据包络分析模型的数码迷彩融合特性评价方法研究

发布时间：2018-02-13 18:41

本文关键词： 兵器科学与技术数码迷彩特征提取超效率数据包络分析模型　出处：《兵工学报》2017年11期 　论文类型：期刊论文

【摘要】：针对活动目标的数码迷彩伪装,为了从光学波段方面评价多种数码迷彩在不同背景中的融合性,对图像内容进行分析,选用HSV颜色模型,分别从图像的色调(H)、饱和度(S)、亮度(V)3个分量提取数码迷彩和背景图像的颜色、纹理、形状、熵及复杂度等5个方面的特征值。根据超效率数据包络分析模型原理,把数码迷彩与背景图像中各个特征差异值作为决策单元(DMU)的输入量,计算出迷彩与背景一对一融合性的效率值。将效率值数据取倒数作为新的DMU输入量,计算得到每种迷彩对不同背景融合性的效率值,从而实现数码迷彩对背景图像综合适应性的评价。利用图像显著性指标对评价结果进行验证,将融合性效率值最大与最小的两种数码迷彩对背景实施伪装,前者在背景图像中的显著性低,具有更好的背景融合性。
[Abstract]:Aiming at the camouflage of digital camouflage of moving target, in order to evaluate the fusion of various digital camouflages in different backgrounds from the optical band, the content of the image is analyzed, and the HSV color model is selected. The color, texture, shape, entropy and complexity of digital camouflage and background image are extracted from the three components of hue, saturation and luminance, respectively. According to the principle of super-efficient data envelope analysis model, the feature values of digital camouflage and background image are extracted. The difference value of each characteristic between digital camouflage and background image is taken as the input of DMU), and the efficiency value of one-to-one fusion of camouflage and background is calculated. The reciprocal of the efficiency value data is taken as the new DMU input quantity. The efficiency value of each camouflage for different background fusion is calculated, and the comprehensive adaptability of digital camouflage to background image is evaluated. The evaluation results are verified by image saliency index. The two kinds of digital camouflage with maximum and minimum fusion efficiency are used to camouflage the background. The former has low significance in background image and has better background fusion.
【作者单位】： 63956部队;北京理工大学光电学院;
【基金】：军内科研计划项目(KD2015070344B12006)
【分类号】：E951.4;TP391.41

【相似文献】