铁路地质判释遥感图像融合算法优选与应用

Selection and Application of Fusion Algorithms for Interpretation of Remote Sensing Images for Railway Geology

研究目的:图像融合是地质遥感判释工作重要的数据处理方法,融合图像的质量直接影响着不良地质判释的准确性。为提高地质遥感判释的精度和质量,研究利用定性和定量评价方法优选适合工程的图像融合算法。研究结论:(1)HPF变换和小波变换在光谱逼真度和纹理清晰度上效果均较好,可作为工程区地质判释遥感数据融合的优选方法;(2)基于HPF变换融合结果,判释工作区内断裂和不良地质信息,判释准确率达80%以上。采用优选的融合算法进行地质遥感判释工作,能够提高不良地质判释的准确度;(3)由于数据源类型、时间和区域等差异,在铁路地质遥感判释工作开展前,应进行图像融合效果评价,优选图像融合算法;(4)该研究成果可应用于工程地质遥感判释中图像融合算法的评价和优选。

Research purposes： The image fusion is an important method for processing the data in interpretation of remote sensing image. In order to improve the accuracy and quality of interpreting the remote sensing image, the study was done on selection of the image fusion algorithms with the qualitative and the quantitative evaluation methods. Research conclusions： （1） The HPF transform and wavelet transform are good in spectral fidelity and texture clarity and they can be used as the preferred data fusion methods for interpretation of the remote sensing image for the engineering geology. （2） Based on the HPF transform result, the information about the fault and unfavorable geological conditions can be interpreted with the accuracy rate above 80 percent. The interpretation accuracy for the unfavorable geological conditions can be improved by using the preferred fusion algorithm. （3） Because of the differences in terms of the data source, time and region, the effect of the fusion algorithms should be evaluated and selected before the image interpretation. （4） This study result can be used for the evaluation and selection of the fusion algorithms.