近年來影像的感測器不斷進步，因此影像的解析度也跟著隨之成長，甚至超過了一億畫素。在此同時也造成了一個問題，那就是在一個影像中包含了過多的資料量，導致需要花費大量的運算時間。而超像素分割是將一群相似顏色或相似結構群組在一起，透過此方法可以大幅降低影像的複雜度。如今超像素演算法在各種領域都被使用，例如深度估計、物件分割、視覺突出物檢測或物件定位等等。超像素演算法在影像處理中常常作為一個前處理的步驟，透過此步驟可以大幅降低影像的複雜度。
在現有的超像素演算法當中常常會面臨到過度分割的問題，在結構簡單的影像中如果過度分割會讓計算的負擔提升。在此篇論文中，我們提出一個以CUDA實現基於色彩及座標距離的超像素迭代分割演算法，此方法透過距離以及色彩相似度來尋找鄰近相似的像素，並且再透過相似度去分析此超像素是否繼續分割下去。接著我們迭代得去執行上述的步驟直到達成了我們所設置的閥值，最後我們在透過轉移超像素的中心點去提升超像素的準確度。透過上述的方法，我們可以大幅減少過度分割的問題並且維持準確度。最後我們透過平行化運算降低此演算法的運算時間，並且達到即時運算。

In recent years, image sensors have continuously progressed, so the resolution of images has also grown, even exceeding 100 million pixels. In the meanwhile, it also caused a problem, that is an image contains too much information, which leads to a lot of computing time. The superpixel segmentation is to construct a group of pixels with similar structures. This algorithm can greatly reduce the complexity of the image. Nowadays, superpixel algorithm are used in various fields, such as depth estimation, object segmentation, visual saliency detection, or object localization, etc. The superpixel algorithm is often used as a pre-processing step in image processing, through this step, the complexity of the image can be greatly reduced.
Existing superpixel algorithms often face the problem of over-segmentation. In an image with simple structure, over-segmentation will increase the computation time. In this thesis, we propose an iterative superpixel segmentation based on color and coordinate distance using CUDA. This algorithm uses distance and color similarity to associate neighboring pixels and then analyzes whether to continue segment the superpixel or not through the similarity. Then we continue the above step iteratively until it reaches the threshold we set. And then we can enhance the accuracy by shifting the superpixel’s center. Through the methods, we can reduce the problem of over-segmentation and maintain accuracy. Finally, we reduce computation time through parallel processing and achieve real-time computing using CUDA.

REFERENCES
[1] C. L. Zitnick and S. B. Kang, "Stereo for image-based rendering using image over-segmentation," in International Journal of Computer Vision (IJCV), 2007.
[2] Z. Li, X. M. Wu and S. F. Chang, "Segmentation using superpixels: A bipartite graph partitioning approach," in IEEE Conference on Computer Vision and Pattern Recognition, 2012.
[3] Y. Li, J. Sun, C. K. Tang and H. Y. Shum, "Lazy snapping," in ACM Transactions on Graphics (SIGGRAPH), 2004.
[4] Z. Liu, X. Zhang, S. Luo and O. Le Meur, "Superpixel-Based Spatiotemporal Saliency Detection," in IEEE Transactions on Circuits and Systems for Video Technology, 2014.
[5] C. Y. Hsu and J. J. Ding, "Saliency detection using DCT coefficients and superpixel-based segmentation," in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2013.
[6] B. Fulkerson, A. Vedaldi and S. Soatto, "Class segmentation and object localization with superpixel neighborhoods," in International Conference on Computer Vision (ICCV), 2009.
[7] D. Xiang, T. Tang, L. Zhao and Y. Su, "Superpixel Generating Algorithm Based on Pixel," in IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2013.
[8] M. Wälivaara, "General Object Detection using Superpixel Preprocessing," Master, Department of Electrical Engineering, Linköping University, 2017
[9] R. Achanta, A. Shaji, K. Smith, A. Lucchi, P. Fua and S. Süsstrunk, "SLIC superpixels compared to state-of-the-art superpixel methods," in IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012.
[10] A. Moore, S. Prince, J. Warrell, U. Mohammed and G. Jones, "Superpixel Lattices," in IEEE Computer Vision and Pattern Recognition (CVPR), 2008
[11] F. Drucker and J. MacCormick, "Fast superpixels for video analysis," in Workshop on Motion and Video Computing, 2009.
[12] P. Siva and A. Wong, "Grid seams: A fast superpixel algorithm for real-time applications," in Proceedings - Conference on Computer and Robot Vision, 2014.
[13] OpenCV Website, [Online], [http://opencv.org]
[14] I. Sobel, "An Isotropic 3x3 Image Gradient Operator," in Stanford A.I. Project, 1968.
[15] J. Canny, "A Computational Approach To Edge Detection, " IEEE Trans. Pattern Analysis and Machine Intelligence, 8(6):679–698, 1986.
[16] X. Ren and J. Malik, “Learning a classification model for segmentation,”in ICCV, 2003, pp. 10–17.
[17] S. Jianbo and J. Malik, "Normalized cuts and image segmentation," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 22, pp. 888-905, 2000.
[18] P. F. Felzenszwalb and D. P. Huttenlocher, "Efficient Graph-Based Image Segmentation," in International Journal of Computer Vision, vol. 59, pp. 167-181, 2004.
[19] M. Y. Liu, O. Tuzel, S. Ramalingam, and R. Chellappa, "Entropy rate superpixel segmentation," in IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2097-2104, 2011.
[20] D. Comaniciu and P. Meer, "Mean shift: a robust approach toward feature space analysis," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 24, pp. 603-619, 2002
[21] A. Levinshtein, A. Stere, K. N. Kutulakos, D. J. Fleet, S. J. Dickinson, and K. Siddiqi, "TurboPixels: Fast Superpixels Using Geometric Flows," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 31, pp. 2290-2297, 2009.
[22] A. Vedaldi and S. Soatto, "Quick Shift and Kernel Methods for Mode Seeking," in 10th European Conference on Computer Vision, pp. 705-718, 2008.
[23] D. Chai, Y. Huang, Y. Bao, “IRSL: Iterative Refining Superpixel Lattice,” IEEE Geoscience and sensing letters, vol. 14, no. 3, pp.344-348, 2017.
[24] S. Avidan and A. Shamir, "Seam carving for content-aware image resizing," in ACM Transactions on Graphics, 2007.
[25] K. M. Laren, "The Development of the CIE 1976 (L* a* b*) Uniform Colour Space and Colour‐difference Formula, "Journal of the Society of Dyers and Colourists, vol. 92, no.9, pp. 317-341, 1976.
[26] ITU-R Recommendation BT.709, International Telecommunication Union - Radiocommunication sector, 1990.
[27] G. Amer and A. Abushaala, "Edge detection methods," World Symposium on Web Applications and Networking (WSWAN), 2015.
[28] A. Kalra, R. L. Chhokar, "A Hybrid Approach Using Sobel and Canny Operator for Digital Image Edge Detection," International Conference on Micro-Electronics and Telecommunication, pp. 305-310, 2016.
[29] S. Dasgupta, C.H. Papadimitriou, and U.V. Vazirani, "Algorithms," pp.163, ISBN: 9780077388492, 2006
[30] S. G. Miaou; J. S. Chou. "Fundamentals of probability and statistics, " 高立圖書, pp.98 ,2012
[31] T. W. Anderson, "An introduction to multivariate statistical analysis, " New York: Wiley, ISBN: 9788126524488, 1958.
[32] J. L. Williams, J. W. Fisher, A. S. Willsky, "Approximate Dynamic Programming for Communication-Constrained Sensor Network Management," IEEE Transactions on Signal Processing, vol.55, no.8, 2007.
[33] J. M. Chang, M. Pedram, "Codex-dp: co-design of communicating systems using dynamic programming," IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol.19, no.7, 2000.
[34] R. W. G. Hunt, M. R. Pointer, “Measuring Colour,” ISBN:9781119975373, 2011.
[35] A. F. M. G. Kibria, M. M. Islam, “Color Image Segmentation Using Visible Color Difference and Canny Edge Detector, ” IEEE Computer and Information Technology, Dec. 2012.
[36] 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 IEEE International Conference on Computer Vision (ICCV), 2001.
[37] B. Y. Wu, "Network Clustering Using Structure Similarity with CUDA Architecture," Master, Department of Electronic and Computer Engineering, National Taiwan University of Science and Technology, 2013.
[38] S. Cook, CUDA Programming: A Developer’s Guide to Parallel Computing with GPUs. Newnes, 2012.
[39] R. Farber, CUDA application design and development. Access Online via Elsevier. 2011.