| ETM与SPOT影像融合方法比较和评价——以庐山及其周边区域为例 |
| |
| 引用本文： | 杨泽宇,韦玉春.ETM与SPOT影像融合方法比较和评价——以庐山及其周边区域为例[J].测绘软科学研究,2011(5):71-76. |
| |
| 作者姓名： | 杨泽宇 韦玉春 |
| |
| 作者单位： | 南京师范大学地理科学学院,江苏南京210046 |
| |
| 基金项目： | 基金项目：国家自然科学基金项目（40771152）资助;江苏省普通高校自然科学研究计划项目（09KJA420001,07KJB4200621资助 |
| |
| 摘 要： | 高空间分辨率影像与多光谱影像融合，为影像处理提供了更加丰富的资源，有利于进行影像的分析和制图。本文以庐山及其周边区域的2009年5月11日LandsatETM＋与2009年5月6日SPOT5影像数据为例，对典型的融合方法进行了对比评价。融合方法包括加权融合法，HIS变换，改进的HIS变换，高频调制融合法（频域采用Butterworth滤波器．指数滤波器），HSV变换，Brovey变换，CN变换，PC变换，Gram—Schmidt变换11种方法。评价指标为相关系数、信息熵、标准差、平均梯度、相对偏差、标准偏差。结果表明，不同的融合结果突出的侧重点不同。加权融合法、Grain—Schmidt变换，PC变换方法突出了影像的信息量丰富度、清晰度，提高空间分辨率能力。CN波谱锐化，改进的HIS变换，BWO高频调制融合，指数高频调制融合方法光谱保真度较好。改进的HIS变换，BWO高频调制融合，CN波谱锐化，指数高频调制融合法在空间分辨率及光谱保真度、清晰度、信息丰富度方面的能力较均衡。
|
| 关 键 词： | 图像融合 高频调制融合 ETM图像 SPOT图像 庐山 |
Image Fusion and Evaluation of ETM and SPOT -A Case of Lushan Region |
| |
| YANG Ze-yu,WEI Yu-chun.Image Fusion and Evaluation of ETM and SPOT -A Case of Lushan Region[J].Research on Sofi Science of Surveying and Mapping,2011(5):71-76. |
| |
| Authors: | YANG Ze-yu WEI Yu-chun |
| |
| Institution: | (Geography Science College, Nanjing Normal University, Nanjing 210046, China) |
| |
| Abstract: | The image fusion of different spatial resolution images, not only provides more abundant information for image processing, but also is conductive to the image analysis and mapping production. In this paper, typical fusion algorithms were compared and evaluated based on Landsat ETM image of May 11, 2009 and PAN SPOT5 image of May 11, 2009, which covers Lushan and the surrounding region. The fusion algorithms used include weighted fusion, IHS transform, improved IHS transform, high frequency modulation fusion method （Butterworth filter, exponent filter）, HSV lransform, Brovey transform, CN transform, PC transforln, and Gram-Schmidt transform. The evaluation parameters used a,＇e correlation coefficient, information entropy, standard deviation, average gradient, relative deviation, and standard deviation. The resuhs showed that different fusion methods had different feature. Some methods such as weighted fusion method, Gram-Schmidt transform, PC transform method highlighted image information-rich degrees, clarity, and space resolution. Other methods such as CN spectrum of sharpening, improved IHS transform, BW0 high frequency modulation fvsion, exponent high frequency modulation fusion could keep the better image spectrum feature. The improved IHS and BWO high frequency inodulation fusion, CN spectrum of sharpening, exponent high frequency modulation fusion keep much more balance. Key words：image fusion; high frequency modulation fusion： ETM; SPOT; Lushan |
| |
| Keywords: | image fusion high frequency modulation fusion： ETM SPOT Lushan |
| 本文献已被 维普 等数据库收录！ |
|
