论文信息 - Research on Matching of the Remote Sensing Image Matching Based on Histogram Invariant Moments and Genetic Algorithms

Research on Matching of the Remote Sensing Image Matching Based on Histogram Invariant Moments and Genetic Algorithms

Based on the histogram invariant moments and Genetic Algorithms, a new remote sensing image matching method is developed to provide robust matching between the remote sensing images having a change in rotation or scale. Image pyramid technology is integrated into the Genetic Algorithms to provide an effective and robust search order. Histogram invariant moment which is invariant to rotation is used as the image feature descriptor. This method has a good performance in accuracy and speed. At the last part of the paper, a Beijing 5m SPOT image is used as the experiment data to show the result.

[1] Melanie Mitchell,et al. Investigation of image feature extraction by a genetic algorithm , 1999, Optics + Photonics.

[2] Ming-Kuei Hu,et al. Visual pattern recognition by moment invariants , 1962, IRE Trans. Inf. Theory.

[3] Egon Dorrer,et al. Automatic image-matching algorithm based on wavelet decomposition , 1994, Other Conferences.

[4] Tarek A. El-Ghazawi,et al. 2-phase GA-based image registration on parallel clusters , 2001, Future generations computer systems.

[5] R. Wong,et al. Scene matching with invariant moments , 1978 .

[6] Feng Gui. An Image Matching Approach Based on the Invariant Feature of Image Histogram , 2000 .

[7] LI Yu-fei. Recognition for sonar image based on improved histogram invariant moments , 2007 .

[8] Huang Wen. Study on the Convergence of the Model Matching to Image on the Basis of GA , 2007 .

[9] Suman K. Mitra,et al. Technique for fractal image compression using genetic algorithm , 1998, IEEE Trans. Image Process..