郭小丹. 基于WT-MB方法的OLI图像融合——以宁夏回族自治区吴忠市的OLI图像和P5图像为例[J]. 华南师范大学学报(自然科学版), 2019, 51(4): 76-85. doi: 10.6054/j.jscnun.2019069
引用本文: 郭小丹. 基于WT-MB方法的OLI图像融合——以宁夏回族自治区吴忠市的OLI图像和P5图像为例[J]. 华南师范大学学报(自然科学版), 2019, 51(4): 76-85. doi: 10.6054/j.jscnun.2019069
GUO Xiaodan. Remote Sensing Image Fusion Based on the WT-MB Method: A Case Study of the Landsat 8 OLI Image and the CBERS-04 P5 Image of Wuzhong City, Ningxia Hui Autonomous Region[J]. Journal of South China Normal University (Natural Science Edition), 2019, 51(4): 76-85. doi: 10.6054/j.jscnun.2019069
Citation: GUO Xiaodan. Remote Sensing Image Fusion Based on the WT-MB Method: A Case Study of the Landsat 8 OLI Image and the CBERS-04 P5 Image of Wuzhong City, Ningxia Hui Autonomous Region[J]. Journal of South China Normal University (Natural Science Edition), 2019, 51(4): 76-85. doi: 10.6054/j.jscnun.2019069

基于WT-MB方法的OLI图像融合——以宁夏回族自治区吴忠市的OLI图像和P5图像为例

Remote Sensing Image Fusion Based on the WT-MB Method: A Case Study of the Landsat 8 OLI Image and the CBERS-04 P5 Image of Wuzhong City, Ningxia Hui Autonomous Region

  • 摘要: 为解决Landsat 8 OLI多光谱图像和CBERS-04 P5全色图像的融合问题,设计了一种基于WT-MB变换的融合方法;以宁夏回族自治区吴忠市的图像融合为案例,从定量评价、目视判读评价和SVM分类精度评价3个方面比较WT-MB变换方法与GS、PCA、MLT、Brovey变换方法的融合效果,然后以总体精度和Kappa系数评判融合图像的分类精度.实验结果表明:WT-MB变换方法在增强融合OLI图像的空间结构信息的同时,更有效地保持原OLI图像的光谱信息,提高了图像分类精度,能够更准确地服务于环境遥感监测.

     

    Abstract: The WT-MB transform image fusion method was designed to solve the integration of the Landsat 8 OLI multispectral image and the CBERS-04 P5 image. The GS, PCA, MLT, and Brovey methods were compared in their adaptation to deal with the fusion of the image of Wuzhong City, Ningxia Hui Autonomous Region. The fusion effects between the WT-MB and other four methods were also compared in terms of quantitative evaluation, visual interpretation evaluation and SVM classification accuracy evaluation. The classification accuracy of fused images was judged by the overall accuracy and Kappa coefficient. In addition to enhancing the space information of the OLI ima-ge, the WT-MB transform fusion method could maintain the spectral information of the OLI image more effectively, improve the accuracy of image classification and serve the environmental remote sensing monitoring more accurately and effectively.

     

/

返回文章
返回