Abstract In this paper, a Shearing Invariant Texture Descriptor (SITD) is proposed, which is a theoretically and computationally simple method based on the Rotation invariant Local Binary Pattern (Rot-LBP) descriptor. In real-world applications using flatbed scanners, such as paper texture fingerprinting, it’s common for a sheet of paper to rotate during the image acquisition process. Because the rotation is usually not based on the paper’s geometrical centre pivot, the produced image is deformed with irregular rotation resulting in shearing transforms. To tackle the shearing problem, the proposed SITD selects a few patterns from the conventional Rot-LBP to achieve either horizontal or vertical invariance. This paper presents the construction of the SITD operators and their performance in recognizing self-developed and standard image datasets, including real paper texture and Outex images, as well as those with distinctive shapes. The images were distorted with only a shearing transform. The self-developed images were distorted manually, while the standard images were distorted by software. The proposed description method achieved up to 100% correctly recognition rate in all the tested datasets based on the horizontal shear invariant operator. In addition to the accurate performance in all the conducted experiments, the operator significantly outperformed the Rot-LBP and another benchmark method, the Shearing Moment Invariant (SMI). The superiority of the descriptor in recognizing different types of patterns demonstrate its ability to be used in applications where the shearing transform is present.

References

[1] Abbadeni N., Texture Representation And Retrieval Using The Causal Autoregressive Model, Journal of Visual Communication and Image Representation, vol. 21, no. 7, pp. 651- 664, 2010.

[2] Buchanan D., Cowburn P., Jausovec V., Petit D., Seem P., Xiong G., Atkinson D., Fenton K., Allwood A., and Bryan M., Forgery: Fingerprinting Documents and Packaging, Nature, pp. 436-475, 2007.

[3] Clarkson W., Weyrich T., Finkelstein A., Heninger N., Halderman A., and Felten E., Fingerprinting Blank Paper Using Commodity Scanners, in Proceeding of 30th IEEE Symposium on Security and Privacy, California, pp. 301-314, 2009.

[4] Garain U. and Halder B., On Automatic Authenticity Verification Of Printed Security Documents, in Proceeding of the 6th Indian Conference on Computer Vision, Graphics, and 4 ,1hsiLBP 4 ,1vsiLBP 4 ,1hsiLBP Shearing Invariant Texture Descriptor from a Local 411 Image Processing, Bhubaneswar, pp. 706-713, 2008.

[5] Gehrke R. and Greiwe A., Multispectral Image Capturing with Foveon Sensors, in Proceeding of International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Rostock, pp. 151-156, 2013.

[6] Guo K., Labate D., Lim W., Weiss G., and Wilson E., Wavelets with Composite Dilations, Electronic research Announcements of the American Mathematical Society, vol. 10, no. 9, pp. 78-87, 2004.

[7] Guo Z., Zhang L., and Zhang D., A Completed Modelling Of Local Binary Pattern Operator For Texture Classification, IEEE Transactions on Image Processing, vol. 19, no. 6, pp. 1657-1663, 2010.

[8] Han R. and Zhang L., Fabric Defect Detection Method Based on Gabor Filter Mask, in Proceeding of IEEE Global Congress on Intelligent Systems, Washington, pp. 184-188, 2009.

[9] Hill D., Batchelor P., Holden M., and Hawkes D., Medical Image Registration, Physics in Medicine and Biology, vol. 46, no. 3, pp.1-45, 2001.

[10] Kenzel W., Rothermel M., Fritsch D. and Haala N., Image Acquisition and Model Selection for Multi-View Stereo, The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Stuttgart, pp.251- 258, 2013.

[11] Kiertscher T., Fischer R., and Vielhauer, C., Latent Fingerprint Detection Using A Spectral Texture Feature, in Proceeding of the 13th ACM Workshop on Multimedia and Security, New York, pp. 27-32, 2011.

[12] Malik F. and Baharudin B., The Statistical Quantized Histogram Texture Features Analysis for Image Retrieval Based on Median and Laplacian Filters in the DCT Domain, The International Arab Journal of Information Technology, vol. 10, no. 6, pp. 616-624, 2013.

[13] Metois E., Yarin P., Salzman N., and Smith R., Fiber Fingerprint Identification, in Proceeding of 3rd Workshop on Automatic Identification, New York, pp. 147-154, 2002.

[14] Nasrudin M., Wahdan O., and Omar K., Irregular Rotation Deformation from Paper Scanning: An investigation, in Proceeding of the International Neural Network Society Winter Conference, Bangkok, pp. 152-161, 2012.

[15] Nordin M. and abdul-Hamid A., Combining Local Binary Pattern and Principal Component Analysis on T-Zone Face Area for Face Recognition, in Proceeding of International Conference on Pattern Analysis and Intelligent Robotics, Kuala Lumpur, pp. 25-30, 2011.

[16] Official Website Of The Pattern Recognition Research Group-National University of Malaysia http://www.ftsm.ukm.my/pr/, Last Visited 2014.

[17] Ojala T., Maenpaa T., Pietikainen M., Viertola J., Kyll nen J., and Huovinen S., Outex-New Framework For Empirical Evaluation of Texture Analysis Algorithms, in Proceeding of 16th International Conference on Pattern Recognition, Quebec, pp. 701-706, 2002.

[18] Ojala T., Pietikainen M., and Maenpaa T., Multiresolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 24, no. 7, pp. 971-987, 2002.

[19] Petrou M. and Sevilla P., Image Processing: Dealing with Texture, Wiley, 2006.

[20] Richards J., Remote Sensing Digital Image Analysis, Springer, 2012.

[21] Shamsuddin M., Higher Order Centralized Scale-Invariants for Unconstrained Isolated Handwritten Digits, PhD Thesis, Universiti Putra Malaysia, Malaysia, 2000.

[22] V cha P., Haindl M., and Suk T., Colour and Rotation Invariant Textural Features Based on Markov Random Fields, Pattern Recognition Letters, vol. 32, no. 6, pp. 771-779, 2011.

[23] Varma M. and Zisserman A., A Statistical Approach to Material Classification Using Image Patch Exemplars, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 31, no. 11, pp. 2032-2047, 2009.

[24] Vera D., Shear Anisotropic Inhomogeneous Besov Spaces in RD, International Journal of Wavelets, Multiresolution and Information Processing, vol. 12, no. 1, 2014.

[25] Wahdan O., Nasrudin M., and Omar K., SITD and Across-Bin Matching for Deformed Images Acquired by Scanners, in Proceeding of IEEE 10th International Colloquium on Signal Processing and its Applications, Kuala Lumpur, pp. 14-19, 2014.

[26] Wang Z. and Yong H., Texture Analysis And Classification With Linear Regression Model Based On Wavelet Transform, IEEE Transactions on Image Processing, vol. 17, no. 8, pp. 1421-1430, 2008.

[27] Wei H., Zhu H., Gan Y., and Shang L., A New LBP in Texture Classification, in Proceeding of 10th International Conference on Intelligent Computing, Taiyuan, pp. 700-705, 2014.

[28] Zhu B., Wu J., and Kankanhalli M., Print Signatures For Document Authentication, in Proceeding of 10th ACM Conference on Computer and Communications Security, Washington, pp. 145-154, 2003.

[29] uni J., Hirota K., and Rosin L., A Hu Moment Invariant as A Shape Circularity 412 The International Arab Journal of Information Technology, Vol. 14, No. 3, May 2017 Measure, Pattern Recognition, vol. 43, no. 1, pp. 47-57, 2010. Omar Wahdan Received the B.Sc. (2008) in computer science from the University of Baghdad and M.Sc. (2011) in Artificial Intelligence from the Universiti Kebangsaan Malaysia (UKM). Currently, he pursues his Ph.D. in the field of texture authentication at the Faculty of Information Science and Technology-UKM. He is also Graduate Research Assistant at the same Faculty. His research interest includes pattern recognition applications where he has authored several publications in this area. Mohammad Nasrudin is an active researcher at the Centre for Artificial Intelligence Technology (CAIT), UKM. He obtained his B. of Business Administration degree majoring in Computer Information System at the Western Michigan University. He received a master s from Universiti Kebangsaan Malaysia (UKM) and a Ph.D. in AI by joining the Malaysia-Imperial College Doctoral Programme. Mohammad's doctoral thesis was a study of image feature construction using the Trace transform, which was guided by the late Prof. Maria Petrou. His current research focuses on document analysis and recognition and metaheuristic optimization. Khairuddin Omar Received his BSc and M.Sc. in Computer Science from Universiti Kebangsaan Malaysia (UKM) in 1986 and 1989, respectively, and Ph.D. in 2000 from Universiti Putra Malaysia. Currently, he is a Professor at the Faculty of Information Science and Technology, UKM. His research interests includes AI, pattern recognition in decision making with uncertainty-Bayesian Reasoning, Neural Networks, Fuzzy Logic, Fuzzy Neural Networks, 2D and 3D image processing, edge detection, thinning, segmentation, feature extraction, texture, resolution, Fourier, Wavelet, etc., with applications to Jawi/Arabic Manuscripts, biometric authentication.