摘要
针对多光谱、高分辨图像融合的评价问题 ,提出了一种利用模糊积分进行综合评价的评价方法 .该评价方法定义了反映融合图像与多光谱图像和高分辨图像之间相关程度的单因素评价指标 ,利用模糊积分定义了相应的综合评价指标 .理论分析和实验结果表明 ,文中提出的评价方法可以用于多光谱、高分辨图像的融合图像评价问题 。
There are several situations that simultaneously require high spatial and high spectral resolution in a single image. This is particularly important in remote sensing. Due to bandwidth limitations, many remote sensing systems have strengths in either spatial or spectral resolution, which is an inevitable tradeoff. One possible solution comes from the field of data fusion. A number of methods have been proposed for merging panchromatic and multispectral data. But it is not solved how to evaluate fusion results properly at present. In this paper, an evaluation method using Sugeno's integral is proposed to evaluate image merged from a high-resolution image and a low-resolution multispectral image. At first, two measures are defined to evaluate multispectral information and spatial resolution information. Multispectral information measure is defined by the correlation coefficient of merged image and multispectral image. Spatial resolution measure is defined by the correlation coefficient of high frequency component of merged image and high-resolution image. Next, total measure using Sugeno's integral is defined based on the above measures. Analysis shows that the evaluation results using Sugeno's integral is sensitive to the change of high weighted measure, which is more consistent with the subjective feeling of human being than that of weighted average method. Evaluation experiments of different fusion image obtained by different methods show that evaluation results are acceptable and is more suitable to the evaluation problem of fusion of a high-resolution image and a low-resolution multispectral image.
出处
《计算机学报》
EI
CSCD
北大核心
2001年第8期815-818,共4页
Chinese Journal of Computers
基金
国防预研基金"多传感器系统数据融合算法研究"(98J6 .6 .1DZ0 114 )资助
