Study On Porphyry Copper Prospecting Information Of Remote Sensing Using SAR Data

Due to the problem that the optical remote sensing in middle-high vegetation coverage area is difficult to directly obtain the information of earth surface, the microwave remote sensing, which has an ability to penetrate vegetation, is utilized to extract the porphyry copper prospecting information of remote sensing, and a remote sensing porphyry copper prospecting model based on SAR data is established.At first, this paper discusses the description models of linear features, uses lineament and ring form, that its geological, visual and statistical significances behind have carried on exposition, to describe linear features in SAR images, and summarizes some statistical methods of linear features. Then the edge detection method for SAR images is improved, a linear features extraction process is designed, and the linear features extraction procedures are realized. Then the program is utilized to extract the linear features in SAR images, and the linear features extraction results is analyzed, so that a lot of meaningful prospecting information can be obtained. Finally, according to the prospecting information, this paper preliminarily establishes a semi-quantitative porphyry copper prospecting model of remote sensing based on SAR data. In this paper, the basic research method is: first, choose some samples from typical mining areas to study; then, using the research conclusion, sum up the index of model, and establish remote sensing prospecting model; finally, use other typical mining areas in the study area, to validate this remote sensing prospecting model.Innovative achievements of this paper is mainly that IROEWA & NMS, a kind of edge detection algorithm for SAR images, is developed based on improving ROEWA operator; under the premise of good edge shape and edge positioning accuracy, false edge has been effectively suppressed, the edge detection effect for SAR images is got a lot of ascension. In addition, based on the researches of predecessors in computational vision, computational theories of vision representation is combined to the development of technology for aided/automatic interpretation of remote sensing, and an application that can extract lineaments and ring forms in the remote sensing images is developed. This program also provides the technical support for the data processing in prospecting information study.