Measured Spectral Data Analysis Of The Typical Ground Object And Remote Sensing Alteration Information Extraction

From the point of view of spectral curve space and spectral feature space, through the analysis on the spectral data of three main ground objects measured in the field, including rocks, soil and vegetation in the two study regions of Tuquan in Inner Mongolia and Tongren in Qinghai, the author found that similar feature distribution form is basically the same for the two study regions in the both spaces, although there are a few differences caused by the geological environment, ecological environment and climate factors in the different regions.The ETM remote sensing data of Xibayanhua in Inner Mongolia and Awa in Qinghai, the promoter region of the two study area were analyzed in the spectral space. According to the analysis, the author found that the image data of the promoter region of the two study area, the same shooting season, the same kind of band combination of the two-dimensional scatter plot shape is basically the same, and the location of the three main ground objects in these two-dimensional scatter plot with their corresponding study area measured in the field spectral data formed two-dimensional scatter plot of the same location. These findings show that the field measurements of spectral data that can guide the distinction of three types of typical objects in the remote sensing image data.On the analysis of two-dimensional scatter plots which formed by the common band combination of spectral data of field measurements to distinguish between three types of typical objects, raised the ratio of band combination method to distinguish between typical objects, and examples demonstrate the feasibility and applicability of this method, which provides a new approach to the classification of surface features.Based on the above research, a mineralization and alteration information extraction model of spectral characteristics analysis based has been proposed, and Xibayanhua in Inner Mongolia and Awa in Qinghai, for example, that the alteration information extraction using this model to good effect.