超像素分割是將一群有相似顏色或結構的像素建構組成而近一步減少影像資訊量，且利於加速後續的應用。在大多數現有的超像素演算法中，它們都需要被指定一個較大且固定的數量給待處理的影像進而確保圖像能被正確分割。而資料集中可能會包含一些結構相對簡單的圖像，這些簡單的圖像可以用少量的超像素來達到相同的分割準確度。我們提出的方法能於簡單的圖像上自動降低超像素的數量且維持相同的分割準確度。能自動降低超像素數量的關鍵是在近一步分割之前我們分析超像素。
我們提出的方法首先將Sobel邊緣偵測以及Canny邊緣偵測結合當成我們的邊緣能量。然後用動態規劃演算法來最佳化我們的邊緣能量，並迭代找出最佳路徑當我們的超像素邊界。最後我們會分析前一迭代產生的超像素大小以及顏色異質性來當作中止分割的條件，因此我們的方法能自動減少超像素的數量。
從我們的實驗結果可以發現，我們可以與先前的演算法比較且達到相同的分割精度，並且在測試數據集中分割2000個超像素內使用更短的時間。從視覺結果可以看出我們可以有效的減少簡單結構圖像的超像素數量且維持分割準確度。

The superpixel segmentation is to construct groups of pixels with a similar color or structure to reduce the amount of information, which can accelerate further processing steps. In most existing superpixel algorithms, a large fixed number of superpixels are assigned to an image to achieve acceptable accuracy. However, this may not be needed for simple structured images that can be constructed by fewer superpixels to achieve the same segmentation accuracy. The proposed method can automatically reduce the number of superpixels for simple structured images and still keep the accuracy. The key point of the proposed method is the analysis of superpixels before further division.
The proposed method combines both Sobel edge detection and Canny edge detection as the edge energy. Then, it iteratively finds the path from the optimized energy by the dynamic programming. Finally, the size and color heterogeneity of superpixels constructed in previous iterations are analyzed to determine early termination conditions, so it can automatically reduce the number of superpixels.
From the experimental results, we can achieve the same segmentation accuracy as the state-of-art previous work, and use less computing time under 2000 superpixels in the testing dataset. It can be observed from the visual results that the algorithm can automatically reduce the number of superpixels for simple structure images and still maintain the accuracy.

[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, " in 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 International Conference on Computer Vision, 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," in 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," in 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, " in 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] D. Martin, C. Fowlkes, and J. Malik., "Learning to detect natural image boundaries using local brightness, color, and texture cues, " in IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004.
[38] 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.