Multi-Scale Stereo Analysis Based on Local Multi-Model Monogenic Image Feature Descriptors

Abstract:

A multi-scale method based on local multi-model monogenic image feature descriptors (LMFD) is proposed to match interest points and estimate disparity map for stereo images. Local multi-model monogenic image features include local orientation and instantaneous phase of the gray monogenic signal, local color phase of the color monogenic signal and local mean colors in the multi-scale color monogenic signal framework. The gray monogenic signal, which is the extension of analytic signal to gray level image using Dirac operator and Laplace equation, consists of local amplitude, local orientation and instantaneous phase of 2D image signal. The color monogenic signal is the extension of monogenic signal to color image based on Clifford algebras. The local color phase can be estimated by computing geometric product between the color monogenic signal and a unit reference vector in RGB color space. Because the proposed feature descriptors contain local geometric, structure and color information, it is robust against noise and brightness change in feature matching and 3D reconstruction. Experiment results on the synthetic and natural stereo images show the performance of the proposed approach.

• Periodical: Advanced Materials Research (Volumes 433-440)
• Pages: 853-859
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.433-440.853
• Online since: January 2012
• Authors: Xiao Qin He, Jin Jun Li, Xiao Yan Li
• Keywords: 2D Monogenic Signal, Disparity Estimation, Imgae Feature Descriptors, Stereo Vision

References

[1] Jinjun Li, Hong Zhao, Xiang Zhou, and Chengying Shi, Robust stereo image matching using a two-dimensional monogenic wavelet transform., Optics Letters, vol. 34(22), 2009, pp.3514-3516, doi: 10. 1364/OL. 34. 003514.

DOI: 10.1364/ol.34.003514

[2] Jinjun Li, Hong Zhao, Qiang Fu and Kejian Jiang, Space-time stereo analysis combining local structure and modulation features in the monogenic wavelet domain., Optics Letters, vol. 35(7), 2010, pp.1049-1051, doi: 10. 1364/OL. 35. 001049.

DOI: 10.1364/ol.35.001049

[3] LennartWietzke , Gerald Sommer, The Signal Multi-Vector, J Math Imaging Vis, vol. 37, 2010, p.132–150, DOI: 10. 1007/s10851-010-0197-3.

[4] John Vince, The geometric product., Geometric algebra for computer graphics, Springer London, 2008, pp.79-124.

[5] Guillaume Demarcq , Laurent Mascarilla and Pierre Courtellemont, The Color Monogenic Signal: A new framework for color image processing., IEEE international conference on image processing (ICIP), 2009, pp.481-484.

DOI: 10.1109/icip.2009.5414353

[6] Felsberg M, Sommer G, The monogenic signal., IEEE Trans Signal Process, vol. 49(12), 2001, pp.3136-3144.

DOI: 10.1109/78.969520

[7] Norbert Kruger, Michael Felsberg, An explicit and compact coding of geometric and structural image information applied to stereo processing., Pattern Recognition Letters, vol. 25, 2004, pp.849-863.

DOI: 10.1016/j.patrec.2004.01.021

[8] Jinjun Li, Hong Zhao, Kejian Jiang, etc., Multiscale stereo analysis based on local-color-phase congruency in the color monogenic signal framework, Optics Letters, vol. 35(13), 2010, pp.2271-2274, doi: 10. 1364/OL. 35. 002272.

DOI: 10.1364/ol.35.002272

[9] H. Bay, A. Ess, T. Tuytelaars, L. V. Gool, Speeded-Up Robust Features., Computer Vision And Image Understanding, Vol. 110, 2008, p.346.

DOI: 10.1016/j.cviu.2007.09.014

[10] Furukawa Yasutaka and Ponce Jean, Accurate, dense, and robust multi-view stereopsis., IEEE conference on computer vision and pattern recognition Vols 1-8, 2007, pp.2118-2125.

DOI: 10.1109/cvpr.2007.383246