Infrared Image Area Correlation Matching Method Based on Phase Congruency

Infrared images have higher noise and lower resolution than visible images. This makes correspondence based on area correlation more difficult to achieve better disparity image. After analyzing the phase congruency transformed image, the paper presents area correlation matching method. First, the images are transformed by phase congruency. Next, the transformed images are matched based on area correlation. The experiment result indicates that the noise of images is restrained and feature is more distinct after phase congruency. The proposed method is more robust to infrared image than area correlation stereo matching method with gray value.

[1]  Peter Kovesi,et al.  Image Features from Phase Congruency , 1995 .

[2]  Shih-Schon Lin.  Review: Extending Visible Band Computer Vision Techniques to Infrared Band Images , 2001 .

[3]  Ze-Nian Li,et al.  Analysis of disparity gradient based cooperative stereo , 1996, IEEE Trans. Image Process..

[4]  M. Nixon,et al.  Temporal phase congruency , 2004, 6th IEEE Southwest Symposium on Image Analysis and Interpretation, 2004..

[5]  Michael Brady,et al.  Saliency, Scale and Image Description , 2001, International Journal of Computer Vision.

[6]  J. Alison Noble,et al.  2D+T acoustic boundary detection in echocardiography , 2000, Medical Image Anal..

[7]  P Kovesi,et al.  Phase congruency: A low-level image invariant , 2000, Psychological research.

[8]  Peter Kovesi,et al.  Edges Are Not Just Steps , 2002 .

[9]  Z. Xiao,et al.  Symmetry Phase Congruency: Feature Detector Consistent with Human Visual System Characteristics , 2003, WAA.

[10]  Peter Kovesi,et al.  Phase Congruency Detects Corners and Edges , 2003, DICTA.

[11]  John S. Zelek,et al.  Dense Surface from Infrared Stereo , 2007, 2007 IEEE Workshop on Applications of Computer Vision (WACV '07).