Research On Recovering And Application Of The Landsat-7 ETM+SLC-Off Image

Landsat-7 spacecraft, which was launched in April 1999 is the seventh global survey satellites of the Landsat Program. Since the Landsat-7 image data has higher quality than before, the image data has been received by 15 satellite ground stations, and has been widely applied in environment assessment and inspection, global change detection, agriculture, geography research, mapping, archeology, layout management and also become the most popular RS data in China. An instrument malfunction occurred onboard Landsat-7 on May 31, 2003. The problem was caused by failure of the Scan Line Corrector (SLC), and the problem appears to be permanent. This problem affects some research field and the recover of the data become quite insistent.This paper was research under the above background. Based on the essential causations of image deformity, I did a deeply research and developed a ETM+ SLC-Off image repair flow. The whole research includes 3 parts: the pre-processing part, research on repair algorithm, compare of different algorithms.The first part involved artificial-selecting based SLC-Off images registration and tassel based auto-selection SLC-Off images registration. Both of them provide a choice when we need to register two SLC-Off images.The second part is the most important. Based on the research of the SLC-Off images repair algorithm used by RSGS(Remote Sensing Ground Station) and the feature of SLC-Off images, I developed some improved algorithms, include Global Differential Histogram Match, Local Differential Histogram Match, Local Histogram Match. Also, I developed another algorithm called Adaptive Local Regression. After a series of experiment, the result turns out that both the Local Histogram Match and Adaptive Local Regression did a good job in image recover.