A new approach for edge detection in noisy images based on the LPGPCA technique

In this study, a new approach to edge detection on images, corrupted with Gaussian and impulsive noise, is presented. The concept, under the decomposition of image with its principal component being an analysis on local pixel grouping for noise suppression, called LPGPCA based denoising, is adopted in order to obtain noiseless gradient maps for edge extraction. As a result, an algorithm has been developed called LPGPCA-ED. Firstly, horizontal and vertical gradient images are computed; then the gradient images are decomposed into a noiseless phase by applying the LPGPCA algorithm. Once a single gradient map has been obtained, a smart nonmaximum suppression operation is carried out to obtain a binary edge map. To show the accuracy of the proposed edge detector objectively, F-measure and PFOM results of the proposed edge detector on images with different Gaussian and impulsive noise are compared with the results of traditional and certain recently published edge detectors. Objectively, the experimental results on RUG and receiver operating characteristic (ROC) curve databases show that our method has better performance than other corresponding edge detectors. Moreover, subjective experiments on a variety of noise contaminated images show that the LPGPCA-ED algorithm is more robust under high level noise conditions, and is able to reveal well-linked lines and also preserve the structural form of a processed image.

PDF

___

[1] Kim KB, Kim JS, Lee S, Choi JS. Fast image registration using pyramid edge images. International Journal of Intelligent Engineering and Systems 2009; 2: 1-8.
[2] Lakshmi A, Rakshit S. Image restoration: an edge detection based regularization. Computer Vision, Pattern Recognition, Image Processing and Graphics (NCVPRIPG); 1517 December 2011; Hubli, Karnataka, India: IEEE. pp. 114-117.
[3] Ma X, Grimson WEL. Edge-based rich representation for vehicle classification. Computer Vision (ICCV); 1721 October 2005; Los Alamitos, CA, USA: IEEE. pp. 1185-1192.
[4] Topal C, Akinlar C. Edge drawing: a combined real-time edge and segment detector. Journal of Visual Communication and Image Representation 2012; 23: 862-872.
[5] Setayesh M, Zhang M, Johnston M. A novel particle swarm optimisation approach to detecting continuous, thin and smooth edges in noisy images. Information Sciences 2013; 246: 28-51.
[6] Canny J. A computational approach to edge detection. IEEE Transactions on Pattern Analysis and Machine Intelligence 1986, 8: 679-698.
[7] Hou Z, Koh T. Robust edge detection. Pattern Recognition 2003, 36: 2083-2091.
[8] Fesharaki M, Hellestrand GR. A new edge detection algorithm based on a statistical approach. Speech, Image Processing and Neural Networks; 1316 April 1994; Hong Kong, China: IEEE. pp. 21-24.
[9] Ren X. Multi-scale improves boundary detection in natural images, 10th European Conference on Computer Vision; 1218 October 2008; Marseille, France: Springer. pp. 533-545.
[10] Bovik AC, Huang TS, Munson DC Jr. Nonparametric tests for edge detection in noise. Pattern Recognition 1986; 19: 209-219.
[11] Kokkinos I. Boundary detection using F-measure, filter-and feature-(F3) boost. 11th European Conference on Computer Vision; 511 September 2010; Heraklion, Crete, Greece. Springer: pp. 650-663.
[12] Lim HD. Robust rank-order test for edge detection in noisy images. Nonparametric Statistics 2006; 18: 333-342.
[13] Padmavasavi K, Kumar NU, Rao EK, Madhavilatha M. Performance evaluation of adaptive statistical thresholding based edge detection using GLCM in wavelet domain under noisy conditions. International Journal on Graphics, Vision and Image Processing 2010; 10: 35-44.
[14] Rakesh RR, Chaudhuri P, Murthy C. Thresholding in edge detection: a statistical approach. IEEE Transactions on Image Processing 2004; 13: 927-936.
[15] Srivastava GK, Verma R, Mahrishi R, Rajesh S. A novel wavelet edge detection algorithm for noisy images. International Conference on Ultra Modern Telecommunications & Workshops; 1214 October 2009; St. Petersburg, Russia: IEEE. pp. 1-8.
[16] Lee J, Haralick R, Shapiro L. Morphologic edge detection. IEEE Journal of Robotics and Automation 1987; 3: 142-156.
[17] Micheli DE, Caprile B, Ottonello P, Torre V. Localization and noise in edge detection. IEEE Transactions on Pattern Analysis and Machine Intelligence 1989; 11: 1106-1117.
[18] Aizenberg IN, Aizenberg NN, Vandewalle J. Precise edge detection: representation by Boolean functions, implementation on the CNN. Fifth IEEE International Workshop on Cellular Neural Networks and Their Applications Proceedings; 1417 April 1998; London, England: IEEE. pp. 301-306.
[19] Gudmundsson M, El-Kwae EA, Kabuka MR. Edge detection in medical images using a genetic algorithm. IEEE Transactions on Medical Imaging 1998, 17: 469-474.
[20] Shui PL, Zhang WC. Noise-robust edge detector combining isotropic and anisotropic Gaussian kernels. Pattern Recognition 2012; 45: 806-820.
[21] Zhang L, Dong W, Zhang D, Shi G. Two-stage image denoising by principal component analysis with local pixel grouping. Pattern Recognition 2010; 43: 1531-1549.
[22] Priya GH, Venkatesh B, Srivani A, Sowmya G, Rajasekhar N. Noise removal in image using LPG-PCA (Local Pixel Grouping-Principle Component Analysis) algorithm. Pattern Recognition 2010; 43: 1531-1549.
[23] Ray R. Unsupervised edge detection and noise detection from a single image. Pattern Recognition 2013; 46: 2067- 2077.
[24] Lopez CM, Baets BD, Bustince H. Quantitative error measures for edge detection. Pattern Recognition 2013; 46: 1125-1139.
[25] Pratt WK. Digital Image Processing. New York, NY, USA: Wiley, 1978.
[26] Von Gioi RG, Jakubowicz J, Morel JM, Randall G. LSD: a line segment detector. Image Processing On Line 2012; 2: 35-55.
[27] Papari G, Petkov N, Campisi P. Artistic edge and corner enhancing smoothing. IEEE Transactions on Image Processing 2007; 16: 2449-2462.
[28] Rafiee G, Dlay S, Woo W. Unsupervised segmentation of focused regions in images with low depth of field. IEEE International Conference on Multimedia and Expo (ICME), 1519 July 2013; San Jose, CA, USA: IEEE. pp. 1-6.
[29] Derraz F, Taleb AA, Beladgham M, Khelif M. Evaluation of segmentation quality for the geometrical active contour method. Image and Signal Processing and Analysis, 1617 September 2005; Zagreb, Croatia: IEEE. pp. 377-382.
[30] Dembczynski K, Jachnik A, Kotlowski W, Waegeman W, Huellermeier E. Optimizing the F-Measure in multi-label classification: plug-in rule approach versus structured loss minimization, Proceedings of the 30th International Conference on Machine Learning; 1621 June 2013; Atlanta, GA, USA: IEEE. pp. 1130-1138.
[31] Kandefer M, Shapiro S. An F-measure for context-based information retrieval, Commonsense 2009; pp. 79-84.
[32] Bowyer K, Kranenburg C, Dougherty S. Edge detector evaluation using empirical ROC curves. IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2325 June 1999; Fort Collins, Colorado, USA:IEEE.
[33] Fern´andez Garc´ıa NL, Carmona Poyato A, Medina Carnicer R, Madrid Cuevas FJ. Automatic generation of consensus ground truth for the comparison of edge detection techniques. http://www.uco.es/ma1fegan/Comunes/investigacion/imagenes/ground-truth.html.
[34] ] Fern´andez Garc´ıa NL, Carmona Poyato A, Medina Carnicer R, Madrid Cuevas FJ. Automatic generation of consensus ground truth for the comparison of edge detection techniques. Image and Vision Computing 2008; 2: 496-511.