臺灣博碩士論文加值系統

影像切割因為其廣泛的應用，像是物體追蹤和影像壓縮，所以在電腦視覺中扮演了非常重要的角色。影像切割是一種將像素聚集成一致且顯著的區域的過程，而且現今已有很多針對不同應用所發展出來的影像切割演算法和技巧。而我們提出來的適應性生長和合併演算法是為了將一張圖分割成使用者所想要的區域數。我們切割方法的流程如下: 首先，產生原始影像的超像素來減少運算量並且提供有用的區域資訊。接著我們使用顏色直方圖和質地來量測兩相鄰超像素的相似度，然後根據該相似度來進行超像素生長，生長的過程會受限於邊緣強度。最後，我們藉由顏色、質地、輪廓、顯著值、以及區域大小替整張圖建立一個相異度矩陣，並按照相異度的大小依序合併區域。合併區域的過程會適應於區域數及影像局部特徵。
經過了超像素生長後，超像素被擴展成比較大的區域，這些大區域擁有更準確的邊緣和區域資訊像是平均顏色和平均質地，有助於最後的區域合併過程。實驗結果顯示我們提出來的方法可以將大部分的圖都切得很好，而且表現還勝過現今較新穎的方法。

In computer vision, image segmentation plays an important role due to its widespread applications such as object tracking and image compression. Image segmentation is a process of clustering pixels into homogeneous and salient regions, and a number of image segmentation algorithms and techniques have been developed for different applications. To segment an image accurately with the number of regions user gives, we propose an adaptive growing and merging algorithm. Our procedure is described as follows: First, a superpixel segmentation is applied to the original image to reduce the computation time and provide helpful regional information. Second, we exploit the color histogram and textures to measure the similarity between two adjacent superpixels. Then we conduct the superpixel growing based on the similarity under the constraint of the edge’s intensity. Finally, we generate a dissimilarity matrix for the entire image according to color, texture, contours, saliency values and region size, and subsequently merge regions in the order of the dissimilarity. The region merging process is adaptive to the number of regions and local image features.
After the superpixel growing has been finished, some superpixels expand to larger regions, which contain more accurate edges and regional information such as mean color and texture, to help with the final process of region merging. Simulations show that our proposed method segments most of images well and outperforms state-of-the-art methods.

口試委員會審定書 #
誌謝 i
中文摘要 ii
ABSTRACT iii
CONTENTS iv
LIST OF FIGURES vii
LIST OF TABLES xii
Chapter 1 Introduction 1
1.1 Motivation 1
1.2 Main Contribution 2
1.3 Organization 2
Chapter 2 Review of Superpixel Segmentation Methods 3
2.1 Mean Shift (MS) 3
2.1.1 Preliminaries 4
2.1.2 The Main Concept of MS 5
2.1.3 The Algorithm 7
2.1.4 Simulations 8
2.2 Normalized Cut (Ncut) 10
2.2.1 The Main Concept of Ncut 10
2.2.2 The Algorithm 12
2.2.3 Simulations 14
2.3 Simple Linear Iterative Clustering (SLIC) 16
2.3.1 The Main Concept of SLIC 16
2.3.2 The Algorithm 19
2.3.3 Simulations 20
2.4 Entropy Rate Superpixel (ERS) 22
2.4.1 The Main Concept of ERS 22
2.4.2 Simulations 26
Chapter 3 Review of Recent Image Segmentation Methods 28
3.1 Multi-Layer Spectral Segmentation (MLSS) 28
3.1.1 The Main Concept to MLSS 28
3.1.2 The Algorithm 31
3.1.3 Simulations 32
3.2 Segmentation by Aggregating Superpixels (SAS) 34
3.2.1 The Main Concept of SAS 34
3.2.2 The Algorithm 36
3.2.3 Simulations 38
3.3 Hierarchical Image Segmentation 40
3.3.1 The Main Concept of OWT-UCM 40
3.3.2 The Algorithm 43
3.3.3 Simulations 44
Chapter 4 Proposed Segmentation Method 46
4.1 Introduction 46
4.2 Superpixel Generation and Saliency Detection 49
4.2.1 Superpixel Generation 49
4.2.2 Saliency Detection 50
4.3 Edge Detection and Texture Features 52
4.3.1 Edge Detection 52
4.3.2 Texture Features 53
4.4 Growing and Merging 55
4.4.1 Superpixel Growing 55
4.4.2 Adaptive Region Merging 56
4.5 Proposed Algorithm 61
4.6 Analysis of Our Algorithm 66
Chapter 5 Simulations 73
5.1 Parameter Setting 73
5.2 Database and Evaluation Metrics 74
5.3 Comparison to the State-of-the-art Methods 76
5.3.1 Comparison of Performance Evaluation 76
5.3.2 Visual Comparison 78
Chapter 6 Conclusion and Future Work 89
6.1 Conclusion 89
6.2 Future Work 90
REFERENCE 91

A.Superpixel
[1]D. Comaniciu and P. Meer, “Mean shift: A robust approach toward feature space analysis,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 24, no. 5, pp. 603-619, May 2002.
[2]J. Shi and J. Malik, “Normalized cuts and image segmentation,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 22, no. 8, pp. 888-905, Aug. 2000.
[3]R. Achanta, A. Shaji, K. Smith, A. Lucchi, P. Fua, and S. Süsstrunk, “SLIC superpixels compared to state-of-the-art superpixel methods,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 34, no. 11, pp. 2274 - 2282, May 2012.
[4]Y. M. Liu, O. Tuzel, S. Ramalingam, and R. Chellappa, “Entropy rate superpixel segmentation,” in CVPR, pp. 2097-2104, 2011.
[5]P. Felzenszwalb and D. Huttenlocher, “Efficient graph-based image segmentation,” Int’l J. Computer Vision, vol. 59, no. 2, pp. 167-181, Sept. 2004.
[6]L. Vincent and P. Soille, “Watersheds in digital spaces: An efficient algorithm based on immersion simulations,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 13, no. 6, pp. 583-598, June 1991.
B.Image Segmentation
[7]P. Arbelaez, M. Maire, C. Fowlkes, and J. Malik, “Contour detection and hierarchical image segmentation,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 33, no. 5, pp. 898-916, May 2011.
[8]P. Arbelaez, M. Maire, C. Fowlkes, and J. Malik, “From contours to regions: An empirical evaluation,” in CVPR, pp.2294-2301, 2009.
[9]T. Kim and K. Lee, “Learning full pairwise affinities for spectral segmentation,” in CVPR, pp. 2101-2108, 2010.
[10]Z. Li, X. M. Wu, and S. F. Chang, “Segmentation using superpixels: A bipartite graph partitioning approach,” in CVPR, pp. 789-796, 2012.
[11]T. Cour, F. Benezit, J. Shi, “Spectral segmentation with multiscale graph decomposition,” in CVPR, pp. 1124-1131, 2005.
[12]P. Arbelaez, “Boundary extraction in natural images using ultrametric contour maps,” in POCV, 2006.
[13]L. Najman and M. Schmitt, “Geodesic saliency of watershed contours and hierarchical segmentation,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 18, no. 12, pp. 1163-1173, Dec 1996.
[14]Y. Deng and B. S. Manjunath, “Unsupervised segmentation of color-texture regions in images and video,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 23, no. 8, pp. 800-810, 2001.
[15]S. R. Rao, H. Mobahi, A. Y. Yang, S. S. Sastry, and Y. Ma, “Natural image segmentation with adaptive texture and boundary encoding,” in ACCV, pp. 135-146, 2009.
[16]C. Y. Hsu and J. J. Ding, “Efficient image segmentation algorithm using SLIC superpixels and boundary-focused region merging,” in ICICS, pp. 1-5, 2013.
[17]D. Martin, C. Fowlkes, D. Tal, and J. Malik. “A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics,” in ICCV, pp. 416–423, 2001.
[18]T. Cour, F. Benezit, and J. Shi, “Spectral segmentation with multiscale graph decomposition,” in CVPR, vol. 2, pp. 1124-1131, 2005.
[19]J. Wang, Y. Jia, X. S. Hua, C. Zhang, and L. Quan, “Normalized tree partitioning for image segmentation,” in CVPR, pp. 1-8, 2008.
[20]M. Donoser, M. Urschler, M. Hirzer, and H. Bischof, “Saliency driven total variation segmentation,” in ICCV, pp. 817-824, 2009.
[21]R. Unnikrishnan, C. Pantofaru, and M. Hebert, “Toward objective evaluation of image segmentation algorithms,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 29, no. 6, pp. 929-944, 2007.
[22]M. Meilǎ, “Comparing clusterings: an axiomatic view,” in ICML, pp. 577-584, Aug. 2005.
[23]D. Martin, C. Fowlkes, D. Tal, and J. Malik, “A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics,” in ICCV, vol. 2, pp. 416-423, 2001.
[24]J. Freixenet, X. Muñoz, D. Raba, J. Martí, and X. Cufí, “Yet another survey on image segmentation: Region and boundary information integration,” in ECCV, pp. 408-422, 2002.
[25]X. Wang, Y. Tang, S. Masnou, and L. Chen, “A Global/Local Affinity Graph for Image Segmentation,” IEEE Trans. Image Processing, vol. 24, no. 4, pp. 1399-1411, 2015.
[26]Y. Yang, Y. Wang, and X. Xue, “A novel spectral clustering method with superpixels for image segmentation,” Optik-International Journal for Light and Electron Optics, vol. 127, no. 1, pp. 161-167, 2016.
[27]X. Gu, J. D. Deng, and M. K. Purvis, “Improving superpixel-based image segmentation by incorporating color covariance matrix manifolds,” in ICIP, pp. 4403-4406, Oct. 2014.
[28]H. Chen, H. Ding, X. He, and H. Zhuang, “Color image segmentation based on seeded region growing with Canny edge detection,” in Signal Processing (ICSP), pp. 683-686, Oct. 2014.
[29]P. R. Narkhede, and A. V. Gokhale, “Color image segmentation using edge detection and seeded region growing approach for CIELab and HSV color spaces, “ in ICIC, pp. 1214-1218, May 2015.
[30]C. C. Kang, W. J. Wang, and C. H. Kang, “Image segmentation with complicated background by using seeded region growing,” AEU-International Journal of Electronics and Communications, vol. 66, no. 9, pp. 767-771, 2012.
[31]F. Y. Shih, and S. Cheng, “Automatic seeded region growing for color image segmentation.” Image and vision computing, vol. 23, no. 10, pp. 877-886, 2005
C.Clustering Techniques
[32]T. Kanungo, D. M. Mount, N. S. Netanyahu, C. D. Piatko, R. Silverman, and A. Y. Wu, “An efficient k-means clustering algorithm: Analysis and implementation,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 24, no. 7, pp. 881-892, 2002.
D.Computer Vision
[33]D. Zhou, O. Bousquet, T.N. Lal, J. Weston, and B. Scholkopf, “Learning with local and global consistency,” Proc. Neural Information Processing Systems, 2003.
[34]S. Beucher and F. Meyer, Mathematical morphology in image processing, Marcel Dekker, 1992, ch. 12.
[35]G. Sharma, W. Wu and E. Dalal, “The CIEDE2000 color-difference formula: Implementation notes, supplementary test data, and mathematical observations,” Color Research & Application, vol. 30, no. 1, pp. 21-30, 2005.
E.Edge Detection
[36]P. Dollar and C. L. Zitnick, “Structured forests for fast edge detection,” in ICCV, pp. 1841–1848, 2013.
[37]P. Acharjya, R. Das, and D. Ghoshal, “A Study on Image Edge Detection Using the Gradients,” International Journal of Scientific and Research Publications, vol. 2, no. 12, Dec. 2012.
[38]D. Ziou and S. Tabbone, “Edge detection techniques-an overview,” Int. J. Pattern Recognit. Image Anal., vol. 8, no.4, pp.537-559, 1998.
F.Saliency Detection
[39]W. Zhu, S. Liang, Y. Wei and J. Sun, “Saliency optimization from robust background detection,” in CVPR, pp. 2814-2821, 2014.
G.Theorems and Mathematics
[40]David J. Field, “Relations between the statistics of natural images and the response properties of cortical cells,” JOSA A, vol. 4, no. 12, pp. 2379-2394, 1987.