Texture Classification based on Spectral Analysis and Haralick Features
     Classificação de Texturas mediante análise espectral e Parâmetros de Haralick

Manuel Blanco Valentin, Clécio Roque de Bom, Márcio P. de Albuquerque, Marcelo P. de Albuquerque, Elisângela L. Faria, Maury D. Correia



In this work we discuss a method to classify a set of texturized images based on the extraction of their Haralick Features. This kind of Classification is capable of providing texture-based measurements (such as contrast or correlation) and use them as main parameters to classify the same type of patterns in other images. In order to improve the classification success ratio a spectral analysis of the textures and, therefore, the use of filters, before the classification step, is proposed here. In this work the classification success has been evaluated using Mean and Canny filters. On the other hand, the Principal Component Analysis is used to optimize the features extracted for the patterns on each image, before introduced into the classifier. With this method the classifying success ratio for the KTH-TIPS subset and 10,000 different permutations was increased --in average-- from 72.28\% to 84.25\%.

Keywords: Haralick Features, Texture Classification, Image Processing, Spectrum Analysis, Principal Component Analysis. 

Neste trabalho é discutido um método de classificação de Texturas baseado na extração de parâmetros de Haralick. Este tipo de classificação é capaz de fornecer medidas de texturas (como por exemplo contraste ou correlação) em padrões presentes em imagens e usá -las para classificar o mesmo tipo de padrões em outras imagens. Com o objetivo de melhorar a taxa de sucesso na classificação foi proposto realizar uma análise espectral e, por tanto, o uso de filtros em passo prévio ao processo de classificação. Este trabalho apresenta uma avaliação do desempenho da classificação por meio do uso dos filtros Média e Canny. Por outro lado, a análise de Componentes Principais foi usada para otimizar os parâmetros extraídos dos padrões nas imagens, antes destas serem introduzidas no classificador. Com este método o sucesso na taxa de classificação para a biblioteca de imagens KTH-TIPS e um conjunto de 10.000 permutações diferentes foi incrementado –em média– de 72.28% para 84.25%.

Palavras chave: Parâmetros de Haralick, Classificação de Texturas, Processamento de Imagens, Análise espectral, PCA



Haralick Features; Texture Classification; Image Processing; Spectrum Analysis; Principal Component Analysis

Texto completo:



Marcos William da Silva Oliveira, Nubia Rosa da Silva, An-toine Manzanera, and Odemir Martinez Bruno. Feature ex-traction on local jet space for texture classification. Physica A: Statistical Mechanics and its Applications, 439:160 – 170, 2015.

Degang Xu, Xiao Chen, Yongfang Xie, Chunhua Yang, and Weihua Gui. Complex networks-based texture extraction and classification method for mineral flotation froth images. Min-erals Engineering, 83:105 – 116, 2015.

Arvind R. Yadav, R.S. Anand, M.L. Dewal, and Sangeeta Gupta. Multiresolution local binary pattern variants based tex-ture feature extraction techniques for efficient classification of microscopic images of hardwood species. Applied Soft Com-puting, 32:101 – 112, 2015.

Suganya Ramamoorthy, R. Kirubakaran, and Rajaram Siva Subramanian. Texture Feature Extraction Using MGRLBP Method for Medical Image Classification. In Suresh, LP and Dash, SS and Panigrahi, BK, editor, ARTIFICIAL INTEL-LIGENCE AND EVOLUTIONARY ALGORITHMS IN ENGI-NEERING SYSTEMS, VOL 1, volume 324 of Advances in In-telligent Systems and Computing, pages 747–753, 2015. Inter-national Conference on Artificial Intelligence and Evolution-ary Algorithms in Engineering Systems (ICAEES), Noorul Is-lam Univ, Noorul Islam Ctr Higher Educ, Kumaracoil, INDIA, APR 22-23, 2014.

Hong Li, Xieping Xu, Buer Qi, Nan Bao, Yaonan Zhang, Hang Sun, Liwei Yu, and Yan Kang. An Effective Feature Extraction Method on Mammograms: A Band Shaped Tex-ture Analysis Based on Iris Filter. JOURNAL OF MED-ICAL IMAGING AND HEALTH INFORMATICS, 4(5):787–792, OCT 2014.

D. Abraham Chandy, J. Stanly Johnson, and S. Easter Selvan. Texture feature extraction using gray level statistical matrix for content-based mammogram retrieval. MULTIMEDIA TOOLS AND APPLICATIONS, 72(2):2011–2024, SEP 2014.

Chao Peng, Jian-Ming Zheng, Xu-Bo Li, Yan-Chao Song, and Jiao-Jiao Shi. Feature extraction on machined surface texture image of tool wear based on fractional brown mo-tion. In Shahhosseini, AM, editor, DESIGN, MANUFACTUR-ING AND MECHATRONICS (ICDMM 2015), pages 706–714. Hebei Univ; Beijing Technol & Business Univ; Chengdu Univ, 2015. International Conference on Design, Manufactur-ing and Mechatronics (ICDMM), Adv Sci Technol & Ind Res Ctr, Wuhan, PEOPLES R CHINA, APR 17-18, 2015.

Satrajit Mukherjee, Bodhisattwa Prasad Majumder, Aritran Piplai, and Swagatam Das. An Adaptive Differential Evolu-tion Based Fuzzy Approach for Edge Detection in Color and Grayscale Images. In Panigrahi, BK and Suganthan, PN and Das, S and Dash, SS, editor, SWARM, EVOLUTIONARY, AND MEMETIC COMPUTING, PT I (SEMCCO 2013), volume 8297 of Lecture Notes in Computer Science, pages 260–273, 2013. 4th International Conference on Swarm, Evolutionary, and Memetic Computing (SEMCCO), SRM Univ, Chennai, INDIA, DEC 19-21, 2013.

Daniel Merkel, Eckard Brinkmann, Joerg C. Kaemmer, Miriam Koehler, Daniel Wiens, and Karl-Michael Derwahl. Comparison Between Various Color Spectra and Conventional Grayscale Imaging for Detection of Parenchymal Liver Le-sions With B-Mode Sonography. JOURNAL OF ULTRA-SOUND IN MEDICINE, 34(9):1529–1534, SEP 2015.

Jing Hu, Daoliang Li, Qingling Duan, Guifen Chen, and Yeiqi Han. Texture Extraction and Analysis by Statistical Methods for Fish Species Classification. SENSOR LETTERS, 11(6-7, SI):1110–1114, JUN-JUL 2013.

Harun Gunes, Elif Nisa Unlu, and Ayhan Saritas. CT versus grayscale rib series for the detection of rib frac-ture. AMERICAN JOURNAL OF EMERGENCY MEDICINE, 33(10):1515–1516, OCT 2015.

R.M. Haralick, K. Shanmugam, and Its’Hak Dinstein. Textural features for image classification. Systems, Man and Cybernetics, IEEE Transactions on, SMC-3(6):610–621, Nov 1973.[13] BlairD. Fleet, Jinyao Yan, DavidB. Knoester, Meng Yao, Jr.

Deller, JohnR., and ErikD. Goodman. Breast cancer detec-tion using haralick features of images reconstructed from ul-tra wideband microwave scans. In Marius George Lingu-raru, Cristina Oyarzun Laura, Raj Shekhar, Stefan Wesarg, Miguel Angel Gonzalez Ballester, Klaus Drechsler, Yoshi-nobu Sato, and Marius Erdt, editors, Clinical Image-Based Procedures. Translational Research in Medical Imaging, vol-ume 8680 of Lecture Notes in Computer Science, pages 9–16. Springer International Publishing, 2014.

Michael V Boland, Mia K Markey, Robert F Murphy, et al. Automated recognition of patterns characteristic of subcellu-lar structures in fluorescence microscopy images. Cytometry, 33(3):366–375, 1998.

Neelamma K Patil, Virendra S Malemath, and Ravi M Yadahalli. Color and texture based identification and classification of food grains using different color models and haralick fea-tures. International Journal on Computer Science and Engi-neering, 3(12):3669, 2011.

Margarete Linek, Matthias Jungmann, Thomas Berlage, Renate Pechnig, and Christoph Clauser. Rock classification based on resistivity patterns in electrical borehole wall images. Journal of Geophysics and Engineering, 4(2):171, 2007.

Xiaofeng Yang, Srini Tridandapani, Jonathan J Beitler, S Yu David, Emi J Yoshida, Walter J Curran, and Tian Liu. Ul-trasound glcm texture analysis of radiation-induced parotid-gland injury in head-and-neck cancer radiotherapy: an in vivo study of late toxicity. Medical physics, 39(9):5732–5739,2012.

M Portes de Albuquerque, IA Esquef, and AR Gesualdi Mello. Image thresholding using tsallis entropy. Pattern Recognition Letters, 25(9):1059–1065, 2004.

Constantino Tsallis. Entropic nonextensivity: a possible measure of complexity. Chaos, Solitons & Fractals, 13(3):371–391, 2002.

Lucas Correia Ribas, Diogo Nunes Goncalves, Jonatan Patrick Margarido Orue, and Wesley Nunes Goncalves. Fractal di-mension of maximum response filters applied to texture analy-sis. PATTERN RECOGNITION LETTERS, 65:116–123, NOV 1 2015.

Igor Pantic, Sanja Dacic, Predrag Brkic, Irena Lavrnja, Tomis-lav Jovanovic, Senka Pantic, and Sanja Pekovic. Discrimi-natory ability of fractal and grey level co-occurrence matrix methods in structural analysis of hippocampus layers. JOUR-NAL OF THEORETICAL BIOLOGY, 370:151–156, APR 7 2015.

Alvaro G. Zuniga, Joao B. Florindo, and Odemir M. Bruno. Gabor wavelets combined with volumetric fractal dimension applied to texture analysis. PATTERN RECOGNITION LETTERS, 36:135–143, JAN 15 2014.

Xiuling Liu, Haiman Du, Guanglei Wang, Suiping Zhou, and Hong Zhang. Automatic diagnosis of premature ventricu-lar contraction based on Lyapunov exponents and LVQ neu-ral network. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE, 122(1):47–55, OCT 2015.

Pierre Soille and Jean-F Rivest. On the validity of fractal dimension measurements in image analysis. Journal of vi-sual communication and image representation, 7(3):217–229,1996.

Alceu Costa. Hausdorff fractal dimension calculation using the box-counting method code for matlab. 2013.

Alan Wolf, Jack B Swift, Harry L Swinney, and John A Vastano. Determining lyapunov exponents from a time series.

Physica D: Nonlinear Phenomena, 16(3):285–317, 1985.

Henry DI Abarbanel, Reggie Brown, and Matthew B Kennel. Local lyapunov exponents computed from observed data. Journal of Nonlinear Science, 2(3):343–365, 1992.

C. Varsakelis and P. Anagnostidis. On the susceptibility of numerical methods to computational chaos and superstability. Communications in Nonlinear Science and Numerical Simu-lation, 33:118 – 132, 2016.

Harry Zhang. The optimality of naive bayes. AA, 1(2):3, 2004.[30] David J Hand and Keming Yu. Idiot’s bayesˆanot so stupid after all? International statistical review, 69(3):385–398, 2001.

Alper Kursat Uysal. An improved global feature selectionscheme for text classification. Expert Systems with Applications, 43:82 – 92, 2016.

Xiang Ji, Soon Ae Chun, Zhi Wei, and James Geller. Twitter sentiment classification for measuring public health con-cerns. SOCIAL NETWORK ANALYSIS AND MINING, 5(1), DEC 2015.

Jie Lu, Khondaker A. Mamun, and Tom Chau. Pattern classification to optimize the performance of Transcranial Doppler Ultrasonography-based brain machine interface. PATTERN RECOGNITION LETTERS, 66:135–143, NOV 15 2015.

Guozhong Feng, Jianhua Guo, Bing-Yi Jing, and Tieli Sun. Feature subset selection using naive Bayes for text classifi-cation. PATTERN RECOGNITION LETTERS, 65:109–115, NOV 1 2015.

Paulo SR Diniz, Eduardo AB Da Silva, and Sergio L Netto. Digital signal processing: system analysis and design. Cambridge University Press, 2010.

Ya-Hui Xiu and Wen-Qing Wu. A novel edge detection algorithm. In Shahhosseini, AM, editor, DESIGN, MANU-FACTURING AND MECHATRONICS (ICDMM 2015), pages 635–640. Hebei Univ; Beijing Technol & Business Univ; Chengdu Univ, 2016. International Conference on Design, Manufacturing and Mechatronics (ICDMM), Adv Sci Tech-nol & Ind Res Ctr, Wuhan, PEOPLES R CHINA, APR 17-18, 2015.

Daniel Tchiotsop, Beaudelaire Saha Tchinda, Rene Tchinda, and Godpromesse Kenne. Edge detection of intestinal par-asites in stool microscopic images using multi-scale wavelet transform. SIGNAL IMAGE AND VIDEO PROCESSING, 9(1, SI):121–134, DEC 2015.

Jaseema Yasmin and Mohamed Sathik. An Improved Iterative Segmentation Algorithm using Canny Edge Detector for Skin Lesion Border Detection. INTERNATIONAL ARAB JOURNAL OF INFORMATION TECHNOLOGY, 12(4):325–332, JUL 2015.

John Canny. A computational approach to edge detection. Pattern Analysis and Machine Intelligence, IEEE Transactions

on, (6):679–698, 1986.[40] Thomas Moeslund. Canny edge detection. Laboratory of Computer Vision and Media Technology, Aalborg University, Denmark, http://www. cvmt. dk/education/teaching/f09/VGIS8/AIP/canny 09gr820. pdf, 2009.

Svante Wold, Kim Esbensen, and Paul Geladi. Proceedings of he multivariate statistical workshop for geologists and geo-chemists principal component analysis. Chemometrics and Intelligent Laboratory Systems, 2(1):37 – 52, 1987.

Herve´ Abdi and Lynne J Williams. Principal component analysis. Wiley Interdisciplinary Reviews: Computational Statistics, 2(4):433–459, 2010.

Eric Hayman, Barbara Caputo, Mario Fritz, and Jan-Olof Eklundh. On the significance of real-world conditions for ma-terial classification. In Computer Vision-ECCV 2004, pages 253–266. Springer, 2004.

Jin Xie, Lei Zhang, Jane You, and Simon Shiu. Effective texture classification by texton encoding induced statistical features. PATTERN RECOGNITION, 48(2):447–457, FEB 2015.[45] Rakesh Mehta and Karen Egiazarian. Texture Classification Using Dense Micro-block Difference (DMD). In Cremers, D and Reid, I and Saito, H and Yang, MH, editor, COMPUTER VISION - ACCV 2014, PT II, volume 9004 of Lecture Notes in Computer Science, pages 643–658. Singapore Tourism Board; Omron; Nvidia; Garena; Samsung; Adobe; ViSenze; Lee Fdn; Morpx; Microsoft Res; NICTA, 2015. 12th Asian Conference on Computer Vision (ACCV), Singapore, SINGAPORE, NOV 01-05, 2014.

Rouzbeh Maani, Sanjay Kalra, and Yee-Hong Yang. Noise ro-ust rotation invariant features for texture classification. PATTERN RECOGNITION, 46(8):2103–2116, AUG 2013.

Mario Fritz, Eric Hayman, Barbara Caputo, and Jan-Olof Ekundh. The kth-tips database, 2004.

Rafael C Gonzalez and Richard E Woods. Digital image processing 3rd edition, 2007.