Dissertation
Dissertation > Industrial Technology > Automation technology,computer technology > Remote sensing technology > Interpretation, identification and processing of remote sensing images > Image processing methods

Based on multi-scale integration of remote sensing image change detection and damage assessment application

Author YuTianChao
Tutor WangWeiXing
School University of Electronic Science and Technology
Course Signal and Information Processing
Keywords the multi scale fusion Markov random field change detection damage evaluation connected component labeling
CLC TP751
Type Master's thesis
Year 2010
Downloads 176
Quotes 0
Download Dissertation

In this paper, from the view of multi-scale fusion, a method appling multi-scale analysis to remote sensing image is proposed for region variation detection. First of all, Wavelet transform is adopted to decompose original images, then change detection algorithm is used to extract change region of different scale images, and finally Markov random field is applied to fuse different scale change detection results. Since Markov random field fusion method takes full account of the correlation between the adjacent pixels and the links of different scale change detection results, the fusion results are more accurate and practical.In this paper, I have researched the methods of evaluating plane, bridge and airstrip damage based on change detection results of images before and after damage.The paper offers a way to recognize the damaged plane, which can recognize additive, lost, and mobile plane objects through plane objects images before and after damage and change detection results. And then the paper marks off different class of damaged plane using damage area ratio, geometry and texture feature change ratio.First, the paper extracts damage bridge information, which includes bridge area, damage area, bridge length, crater number, max crater area, min crater area, mean crater area, and truncation detail. Then computing damage area ratio, crater number and area per unit length, and truncation region number, the paper makes sure bridge damage degree through these damage guideline parameters.The paper judges airstrip damage, through deciding if rectangle region satisfying plane takeoff exists in the damaged airstrip. First we obtain the damaged airstrip using the airstrip object and change detection results. Then we consider the rectangle region as a template. We spin template in different airstrip position to find the rectangle region satisfying plane takeoff. If we can’t find any one in all airstrip position from all angles, the airstrip is proved damage. If not, we locate the position of the rectangle region.

Related Dissertations
More Dissertations