The Research And Application Of Remote Sensing Alteration Information Extraction Of Vegetation Coverage Area

Abstract:Vegetation coverage area is one of the difficulties of the remote sensing alteration information extraction. This thesis uses Wuzhengdao as the study area, which is located in north of Guizhou province, and achieves satisfactory results with ETM multi-spectrum data. Firstly, it implements preprocessing and enhancement processing to remote sensing image. Secondly, combining with the basic geological data, it carries on the remote sensing geological interpretation. Thirdly, ferric contamination anomaly and hydroxylation anomaly were extracted with remote sensing image by using the combining method of suppressing middle and low vegetation coverage area, masking lush vegetation area and other interference factors area, principal component analysis and density segmentation. Finally, through the studying of geological condition and metallogenic regularity, it implements remote sensing synthetic metallogenic prediction, and points out several prospect-areas. This thesis mainly achieves the following results.1. Through image preprocessing and enhancement processing, remote sensing radiation error and geometric distortion were solved, which improves the clarity of the image texture and is conducive to visual interpretation of remote sensing. This thesis summarizes the features of some kinds of strata and structure by the analysis of the basic geological data available, and gets the interpretation results of strata and structure.2. Interference factors which influence the result of the alteration information extraction mainly include vegetation, water system, shadow and quaternary deposits, etc. This thesis uses normalized difference vegetation index (NDVI) to identify lush vegetation area with middle and low vegetation coverage area. For middle and low vegetation coverage area, vegetation information was suppressed by vegetation suppression algorithm. For the lush vegetation area, water system, shadow and quaternary deposits, this method which contains building each one mask, multiplying each mask and achieving the final mask is very effective to remove the interference factors.3. Ferric contamination anomaly and hydroxylation anomaly were extracted with remote sensing image by using the method of principal component analysis, which were divided into three anomaly level and made into ferric contamination anomaly map and hydroxylation anomaly map. Through the overlay analysis of alteration anomaly and known mineral deposit distribution, the author finds the fact that they have a very good coincidence, which illustrates the fact that ferric contamination anomaly and hydroxylation anomaly which were extracted reflect the mineralization alteration information very well.4. Through the comprehensive analysis of the geological interpretation results, alteration information extraction results, known mineral deposit (points) distribution, regional geological condition and metallogenic regularity, this thesis implements remote sensing synthetic metallogenic prediction, and points out several prospect-areas.