Multi-channel versus quaternion orthogonal rotation invariant moments for color image representation

Abstract Orthogonal rotation invariant moments (ORIMs) have been used in many pattern recognition and image processing applications in the last three decades. Most of the applications relate to monochrome and gray-scale images. Recently, the theory of image moments for gray-scale images has been extended to color images using quaternion moments to explore the benefit of color information while representing the color images by moments. In this paper, we propose multi-channel ORIMs (MORIMs) invariants for color images and compare their performance with the existing quaternion moments, called quaternion orthogonal rotation invariant moments (QORIMs). The theoretical and experimental analysis demonstrates the superiority of the proposed MORIMs over the QORIMs invariants in the color image recognition task. The experiments are conducted by considering Zernike moments (ZMs) and quaternion ZMs (QZMs) as the representatives of MORIMs and QORIMs, respectively.

[1]  Zhiwen Liu,et al.  Quaternion generic Fourier descriptor for color object recognition , 2015, Pattern Recognit..

[2]  Antonio Torralba,et al.  Modeling the Shape of the Scene: A Holistic Representation of the Spatial Envelope , 2001, International Journal of Computer Vision.

[3]  Chandan Singh,et al.  Accurate calculation of Zernike moments , 2013, Inf. Sci..

[4]  Jan Flusser,et al.  Affine Moment Invariants of Color Images , 2009, CAIP.

[5]  Hajo Holzmann,et al.  Testing for Image Symmetries—With Application to Confocal Microscopy , 2009, IEEE Transactions on Information Theory.

[6]  Raveendran Paramesran,et al.  On the computational aspects of Zernike moments , 2007, Image Vis. Comput..

[7]  Zen Chen,et al.  A Zernike Moment Phase-Based Descriptor for Local Image Representation and Matching , 2010, IEEE Transactions on Image Processing.

[8]  Pooja,et al.  Local and global features based image retrieval system using orthogonal radial moments , 2012 .

[9]  Qiang Chen,et al.  A moment-based nonlocal-means algorithm for image denoising , 2009, Inf. Process. Lett..

[10]  Gang Chen,et al.  Color Image Analysis by Quaternion-Type Moments , 2014, Journal of Mathematical Imaging and Vision.

[11]  Xiangyang Wang,et al.  Quaternion polar complex exponential transform for invariant color image description , 2015, Appl. Math. Comput..

[12]  Ekta Walia,et al.  Rotation invariant complex Zernike moments features and their applications to human face and character recognition , 2011 .

[13]  Whoi-Yul Kim,et al.  Robust Rotation Angle Estimator , 1999, IEEE Trans. Pattern Anal. Mach. Intell..

[14]  Jamshid Shanbehzadeh,et al.  Fast Zernike wavelet moments for Farsi character recognition , 2007, Image Vis. Comput..

[15]  Jan Flusser,et al.  Near infrared face recognition by combining Zernike moments and undecimated discrete wavelet transform , 2014, Digit. Signal Process..

[16]  James Ze Wang,et al.  SIMPLIcity: Semantics-Sensitive Integrated Matching for Picture LIbraries , 2001, IEEE Trans. Pattern Anal. Mach. Intell..

[17]  Glenn Healey,et al.  Using Zernike moments for the illumination and geometry invariant classification of multispectral texture , 1998, IEEE Trans. Image Process..

[18]  Bo Yang,et al.  Near infrared face recognition using Zernike moments and Hermite kernels , 2015, Inf. Sci..

[19]  Gang Chen,et al.  Color Image Analysis by Quaternion Zernike Moments , 2010, 2010 20th International Conference on Pattern Recognition.

[20]  Alireza Khotanzad,et al.  Invariant Image Recognition by Zernike Moments , 1990, IEEE Trans. Pattern Anal. Mach. Intell..

[21]  Wang Xiang-yang,et al.  Invariant quaternion radial harmonic Fourier moments for color image retrieval , 2015 .

[22]  Karthik Ramani,et al.  Classifier combination for sketch-based 3D part retrieval , 2007, Comput. Graph..

[23]  Dimitris A. Karras,et al.  A new class of Zernike moments for computer vision applications , 2007, Inf. Sci..

[24]  Raveendran Paramesran,et al.  New computational methods for full and subset Zernike moments , 2004, Inf. Sci..

[25]  Xuelong Li,et al.  Zernike-Moment-Based Image Super Resolution , 2011, IEEE Transactions on Image Processing.

[26]  Pooja,et al.  Improving image retrieval using combined features of Hough transform and Zernike moments , 2011 .

[27]  Ming Zhu,et al.  Quaternion moment and its invariants for color object classification , 2014, Inf. Sci..

[28]  Jan Flusser,et al.  Rotation and noise invariant near-infrared face recognition by means of Zernike moments and spectral regression discriminant analysis , 2013, J. Electronic Imaging.

[29]  Gang Chen,et al.  Quaternion Zernike moments and their invariants for color image analysis and object recognition , 2012, Signal Process..

[30]  Shan Li,et al.  Complex Zernike Moments Features for Shape-Based Image Retrieval , 2009, IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans.

[31]  Ming Zhu,et al.  Quaternion Fourier-Mellin moments for color images , 2011, Pattern Recognit..

[32]  M. Teague.  Image analysis via the general theory of moments , 1980 .

[33]  Dimitris E. Koulouriotis,et al.  A Unified Methodology for Computing Accurate Quaternion Color Moments and Moment Invariants , 2014, IEEE Transactions on Image Processing.

[34]  Atilla Baskurt,et al.  Improving Zernike Moments Comparison for Optimal Similarity and Rotation Angle Retrieval , 2009, IEEE Trans. Pattern Anal. Mach. Intell..