臺灣博碩士論文加值系統

本論文提出一基於完形理論之迭代指向性方法(entropy-driven gradient evaluation, P-EdGE)，其以亂度驅策之評估方式對影像中的灰階梯度方向進行分析，藉此偵測人眼所感受到的物件邊緣(即感知邊緣)。P-EdGE的特徵在於迭代地改變一取樣遮罩以獲得目標像素及鄰近像素之梯度方向，該取樣遮罩終將收斂並覆蓋最適合表現目標像素方向性之鄰近像素。吾人展示此迭代程序中亂度驅策之設計可滿足完形理論(Gestalt theory)中的相似性與鄰近性法則。迭代至收斂後之方向性與目標像素上的梯度強度會經由貝式程序(Bayesian process)計算，其可有效地決定該目標像素是否屬於一感知邊緣。實驗結果顯示， P-EdGE在偵測感知邊緣上相較於其他方法有顯著的優勢。

This thesis presents a novel method based on entropy-driven gradient evaluation (P-EdGE) for detecting perceptual edges that constitute boundaries of objects as perceived by human eyes. P-EdGE is characterized by iteratively employing a shape-changeable mask centered at a target pixel to sample gradient orientations of neighboring pixels for measuring the directivity of the target pixel. The mask deforms to automatically cover pixels most suitable for exhibiting the directivity of the target pixel. We show that such an iterative scheme satisfies the similarity and proximity laws in Gestalt theory. After convergence, the converged directivity in conjunction with the gradient magnitude are subjected to a Bayesian process for determining whether the target pixel belongs to a perceptual edge. Experimental results are presented to justify the superiority of P-EdGE over other methods in detecting perceptual edges.

Table of content

Chinese abstract…………………………………………………………………………………………………………………I
English abstract………………………………………………………………………………………………………………II
Table of content……………………………………………………………………………………………………………III
List of figures…………………………………………………………………………………………………………………IV
List of tables……………………………………………………………………………………………………………………IV
1. INTRODUCTION………………………………………………………………………………………………………1
1.1 Visual Perception…………………………………………………………………………………………1
1.2 Conventional Edge Detectors………………………………………………………………2
1.3 Perceptual Edge in Applications……………………………………………………3
1.4 Gestalt Laws………………………………………………………………………………………………………5
2. THREE POSTULATES ON VISUAL PERCEPTION……………………………………9
2.1 Postulate 1: Noticing Neighboring Pixels with Similar Orientations………………………………………………………………………………………………………9
2.2 Postulate 2: Evaluating Orientations with Large Receptive Field……………………………………………………………………………………………………………………9
2.3 Postulate 3: Perceiving in a Pixel-Based Fashion………………………………………………………………………………………………………………………………………10
3. THE PROPOSED METHOD…………………………………………………………………………………11
3.1 Selection of Pixels of Interest…………………………………………………11
3.2 Entropy-Driven Scheme……………………………………………………………………………12
3.3 Bayesian Decision…………………………………………………………………………………………………18
4. EXPERIMENTS………………………………………………………………………………………………………20
4.1 Experimental Setup……………………………………………………………………………………20
4.2 Experimental Results………………………………………………………………………………20
4.3 Comparison with Other Methods………………………………………………………22
5. CONCLUSION…………………………………………………………………………………………………………24
6. REFERENCES…………………………………………………………………………………………………………25
APPENDIX

[1] Cai, J., Miklavcic, S.: 'Surface fitting for individual image thresholding and beyond', IET Image Process., 2013, 7, (6), pp. 596-605.

[2] Shotton, J., Blake, A., Cipolla, R.: 'Multiscale categorical object recognition using contour fragments', IEEE Trans. Pattern Anal. Mach. Intell., 2008, 30, (7), pp. 1270-1281.

[3] Sánchez-Nielsen, E., Hernández-Tejera, M.: 'Real-time tracking using A* heuristic search and template updating', IET Comput. Vis., 2011, 5, (3), pp. 169-177.

[4] Cai, L., He, L., Yamashita, T., et al.: 'Robust contour tracking by combining region and boundary information', IEEE Trans. Circuits Syst. Video Technol., 2011, 21, (12), pp. 1784-1794.

[5] Roberts, L.G.: 'Machine perception of three-dimensional solids ', Optical and Electro-Optical Inf. Process., Tippett, J.T., et al., eds., MIT Press, 1965.

[6] Treisman, A.M., Gelade, G.: 'A feature-integration theory of attention', Cog. Psychol., 1980, 12, (1), pp. 97-136.

[7] Malik, J.: 'Visual grouping and object recognition ', IEEE 11th Int. Conf. Image Analysis and Processing, 2006, pp. 612-621.

[8] Ullman, S.: 'Filling-in the gaps: the shape of subjective contours and a model for their generation', Biol. Cybern., 1976, 25, (1), pp. 1-6.

[9] Ellis, W. D.: 'A sourcebook of Gestalt psychology' (Psychology Press, United Kingdom, 1999).

[10] Sajda, P., Pohlmeyer, E., Wang, J., et al.: 'In a blink of an eye and a switch of a transistor: cortically coupled computer vision', Proc. IEEE, 2010, 98, (3), pp. 462-478.

[11] Ren, X., Fowlkes, C. C., Malik, J.: 'Learning probabilistic models for contour completion in natural images', Int. J. Comput. Vis., 2008, 77, (1-3), pp. 47-63.

[12] Gonzalez, R.C., Woods, R.E., 'Digital image processing' (Prentice Hall Press, New Jersey, 3rd edn. 2007).

[13] Canny, J.: 'A computational approach to edge detection', IEEE Trans. Pattern Anal. Mach. Intell., 1986, (6), pp. 679-698.

[14] Goldstein, E.: 'Sensation and perception' (Cengage Learning Press, Kentucky, 9th edn. 2013).

[15] Foresti, G.L.: 'Visual inspection of sea bottom structures by an autonomous underwater vehicle', IEEE Trans. Syst., Man, Cybern., C, Cybern., 2001, 31, (5), pp. 691-705.

[16] Kim, A., Eustice, R.: 'Pose-graph visual SLAM with geometric model selection for autonomous underwater ship hull inspection', IEEE/RSJ Int. Conf. Intelligent Robots and Systems, Missouri, USA, Oct. 2009, pp. 1559-1565.

[17] Galceran, E., Campos, R., Palomeras, N., et al.: 'Coverage path planning with real-time replanning and surface reconstruction for inspection of three-dimensional underwater structures using autonomous underwater vehicles', J. Field Robot., 2014.

[18] Mittleman, J., Swan, L.: 'Underwater inspection for welding and overhaul', Naval Engineers J., 1993, 105, (5), pp. 37-42.

[19] Negahdaripour, S., Firoozfam P.: 'An ROV stereovision system for ship hull inspection', IEEE J. Ocean. Eng., 2006, 31, (3), 551-546.

[20] Kunz, C., Singh, H.: 'Map building fusing acoustic and visual information using autonomous underwater vehicles', J. Field Robot., 2013, 30, (5), 763-783.

[21] Mallios, A.: 'Sonar scan matching for simultaneous localization and mapping in confined underwater environments', Ph. D. Thesis, Department of Architecture and Construction Engineering, University of Girona, Spain, 2014.

[22] Quidu, I., Jaulin, L., Bertholom, A., et al.: 'Robust multitarget tracking in forward-looking sonar image sequences using navigational data', IEEE J. Ocean. Eng., 37, (3), 417-430.

[23] Yufit, G., Maillard , E. P.: '3D forward looking sonar technology for surface ships and AUV: Example of design and bathymetry application', IEEE Int. Underwater Technology Symposium, Tokyo, Japan, 2013, pp. 1-5.

[24] Bazeille, S., Quidu, I., Jaulin, L., et al.: 'Automatic underwater image pre-processing', Proceedings of the SEA TECH WEEK Caractérisation du Milieu Marin, 2006.

[25] Elder, J.H., Goldberg, R.M.: 'Ecological statistics of gestalt laws for the perceptual organization of contours', J. Vis., 2002, 4, (2), article 5.

[26] Velidou, D. S., Tolpekin, V. A., Stein, A. et al.: 'Use of Gestalt Theory and Random Sets for Automatic Detection of Linear Geological Features', Math. Geosci., 2015, 47, (3), pp.249-276.

[27] Tseng, C. S., Lin, C. T., Lin, C. W., et al.: 'Perceptual edges preservation conformal to human vision perception', IEEE 13th Int. Conf. Information Reuse and Integration, Las Vegas, U.S.A., Aug. 2012, pp. 173-178.

[28] Desolneux, A., Moisan, L., Morel, J.: 'From gestalt theory to image analysis: a probabilistic approach' (Springer-Verlag Press, New York, 2007).

[29] Von Gioi, R. G., Jakubowicz, J., Morel, J.M., et al.: 'LSD: A fast line segment detector with a false detection control', IEEE Trans. Pattern Anal. Mach. Intell., 2010, 32, (4), pp. 722-732.

[30] Laplace, P.S.: 'Théorie analytique des probabilités' (Ve. Courcier, Paris, 2nd edn. 1814).

[31] Müller-Lyer, F. C.: ' Optische urteilstäuschungen ', Archiv für Anatomie und Physiologie, 1889, (2), pp. 263-270.

[32] Carrasco, M., McElree, B.: 'Covert attention accelerates the rate of visual information processing', Proc. Natl. Acad. Sci. U.S.A., 2001, 98, (9), pp. 5363-5367.

[33] Yeshurun, Y., Carrasco, M.: 'Spatial attention improves performance in spatial resolution tasks', Vision Res., 1999, 39, (2), pp. 293-306.

[34] Itti, L., Koch, C.: 'Computational modelling of visual attention', Nat. Rev. Neurosci., 2001, 2, (3), pp. 194-203.

[35] McAndrew, A., Wang, J.H., Tseng, C.S.: 'Introduction to Digital Image Processing with MATLAB' (Cengage Learning Press, Taipei, Asia ed. 2010).

[36] King, D.B., and Wertheimer, M.: 'Max Wertheimer and Gestalt Theory' (Transaction Publishers Press, New Jersey. 2005).

[37] Hopfield, J.J.: 'Neural networks and physical systems with emergent collective computational abilities', Proc. Natl. Acad. Sci. U.S.A., 1982, 79, (8), pp. 2554-2558.

[38] Gideon, G., Medioni, G.: 'Perceptual grouping using global saliency-enhancing operators', Proc. Int. Conf. Pattern Recogn., The Hague, Holland, 1992, pp. 99-104.

[39] Grigorescu, C., Petkov, N., Westenberg, M. A.: 'Contour and boundary detection improved by surround suppression of texture edges', Image Vision Comput., 2004, 22, (8), pp. 609-622.

[40] Papari, G., Petkov, N.: 'An improved model for surround suppression by steerable filters and multilevel inhibition with application to contour detection', Pattern Recogn., 2011, 44, (9), pp. 1999-2007.

[41] Lopez-Molina, C., De Baets, B., Bustince, H.: ' Quantitative error measures for edge detection ', Pattern Recognition, 2013, 46, (4), pp. 1125-1139.