Fast and accurate computation of orthogonal moments for texture analysis

被引:27
作者
Di Ruberto, Cecilia [1 ]
Putzu, Lorenzo [2 ]
Rodriguez, Giuseppe [1 ]
机构
[1] Univ Cagliari, Dept Math & Comp Sci, Via Osped 72, I-09124 Cagliari, Italy
[2] Univ Cagliari, Dept Elect & Elect Engn, Piazza dArmi, I-09123 Cagliari, Italy
关键词
Texture descriptor; Moment; Local binary pattern; Co-occurrence matrix; Classification; IMAGE-ANALYSIS; FEATURES; CLASSIFICATION; RECOGNITION; KRAWTCHOUK; INVARIANTS; PATTERN; SCALE;
D O I
10.1016/j.patcog.2018.06.012
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In this work we describe a fast and stable algorithm for the computation of the orthogonal moments of an image. Indeed, orthogonal moments are characterized by a high discriminative power, but some of their possible formulations are characterized by a large computational complexity, which limits their real-time application. This paper describes in detail an approach based on recurrence relations, and proposes an optimized Matlab implementation of the corresponding computational procedure, aiming to solve the above limitations and put at the community's disposal an efficient and easy to use software. In our experiments we evaluate the effectiveness of the recurrence formulation, as well as its performance for the reconstruction task, in comparison to the closed form representation, often used in the literature. The results show a sensible reduction in the computational complexity, together with a greater accuracy in reconstruction. In order to assess and compare the accuracy of the computed moments in texture analysis, we perform classification experiments on six well-known databases of texture images. Again, the recurrence formulation performs better in classification than the closed form representation. More importantly, if computed from the GLCM of the image using the proposed stable procedure, the orthogonal moments outperform in some situations some of the most diffused state-of-the-art descriptors for texture classification. (C) 2018 Elsevier Ltd. All rights reserved.
引用
收藏
页码:498 / 510
页数:13
相关论文
共 50 条
[1]   3D image analysis by separable discrete orthogonal moments based on Krawtchouk and Tchebichef polynomials [J].
Batioua, Imad ;
Benouini, Rachid ;
Zenkouar, Khalid ;
Zahi, Azeddine ;
Hakim, El Fadili .
PATTERN RECOGNITION, 2017, 71 :264-277
[2]  
Benco M, 2007, RADIOENGINEERING, V16, P64
[3]   Rotation invariant co-occurrence features based on digital circles and discrete Fourier transform [J].
Bianconi, Francesco ;
Fernandez, Antonio .
PATTERN RECOGNITION LETTERS, 2014, 48 :34-41
[4]   N-FOLDED SYMMETRIES BY COMPLEX MOMENTS IN GABOR SPACE AND THEIR APPLICATION TO UNSUPERVISED TEXTURE SEGMENTATION [J].
BIGUN, J ;
DUBUF, JM .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1994, 16 (01) :80-87
[5]   Scene Classification Based on Gray Level-Gradient Co-Occurrence Matrix in the Neighborhood of Interest Points [J].
Chen, Shuo ;
Wu, Chengdong ;
Chen, Dongyue ;
Tan, Wenjun .
2009 IEEE INTERNATIONAL CONFERENCE ON INTELLIGENT COMPUTING AND INTELLIGENT SYSTEMS, PROCEEDINGS, VOL 4, 2009, :482-485
[6]   Translation and scale invariants of Legendre moments [J].
Chong, CW ;
Raveendran, P ;
Mukundan, R .
PATTERN RECOGNITION, 2004, 37 (01) :119-129
[7]   IMAGE CHARACTERIZATIONS BASED ON JOINT GRAY LEVEL RUN LENGTH DISTRIBUTIONS [J].
DASARATHY, BV ;
HOLDER, EB .
PATTERN RECOGNITION LETTERS, 1991, 12 (08) :497-502
[8]   Palmprint Authentication using Modified Legendre Moments [J].
Deepika, Lakshmi C. ;
Kandaswamy, A. ;
Vimal, C. ;
Satish, B. .
PROCEEDINGS OF THE INTERNATIONAL CONFERENCE AND EXHIBITION ON BIOMETRICS TECHNOLOGY, 2010, 2 :164-172
[9]   Stable, fast computation of high-order Zernike moments using a recursive method [J].
Deng, An-Wen ;
Wei, Chia-Hung ;
Gwo, Chih-Ying .
PATTERN RECOGNITION, 2016, 56 :16-25
[10]  
Di Ruberto C, 2005, LECT NOTES COMPUT SC, V3617, P212, DOI 10.1007/11553595_26