在電腦視覺中，影像分割指的是將一張圖片分割成許多圖像子區域的過程。影像分割最主要的目的是將輸入的圖片轉換成比要有意義的形式，如此一來我們可以更容易對該圖片做分析。近年來，有許多影像分割的方法被提出，而且其應用也相當的廣泛，例如：醫學影像、物體偵測、人臉辨識…。一般來說，影像分割方法可以分成監督式和非監督式兩種。監督式影像分析的結果會受到使用者很大的影響，因此我們提出了一個非監督式的影像分割方法，它不會受到使用者的影響。我們的方法是利用顯著性檢測標示出圖片中重要和不重要的部分，還有利用邊緣檢測標示出圖片中細節的部份，如此一來就可以自動標出前景/背景種子，然後再利用它分割出前景和背景。根據結果顯示，我們的方法效果很好。

In computer vision, image segmentation is the process of partitioning an image into several segments. The goal of image segmentation is to transform the input image into a more meaningful form which is easier to analyze. Nowadays, there are many segmentation approaches, and they can be applied in many fields, such as medical imaging, object detection, face recognition, etc. Generally, image segmentation can be distinguished as supervised and unsupervised two categories. The result of supervised image segmentation is greatly affected by the user. Therefore, we propose an unsupervised method of image segmentation, which can’t be affected by users. We use saliency detection to label some informative and insignificant parts of the image, and then, we apply edge detection to label some details of the image. In this way, we can automatically label the seeds to get the scribble; then segment foreground/background. The results show that our method is good.

[1] L. Grady. Random walks for image segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 28(11):1768–1783, 2006.
[2] P. Kohli, L. Ladicky, and P. H. S. Torr. Robust higher order potentials for enforcing label consistency. IJCV, 82:302–324, 2009.
[3] T. H. Kim, K. M. Lee, and S. U. Lee. Nonparametric higher-order learning for interactive segmentation. CVPR, 3201–3208, 2010.
[4] H. Jiang, J. Wang, Z. Yuan, T. Liu, and N. Zheng. Automatic salient object segmentation based on context and shape prior. In BMVC, 110.1–110.12, 2011.
[5] S. Goferman, L. Zelnik-Manor, and A. Tal. Context-aware saliency detection. CVPR, 2376–2383, 2010.
[6] K. Y. Chang, T. L. Liu, H. T. Chen, and S. H. Lai, "Fusing Generic Objectness and Visual Saliency for Salient Object Detection," in IEEE International Conference on Computer Vision, 2011, pp. 914-921.
[7] C. Yang, L. Zhang, H. Lu, X. Ruan, and M.-H. Yang. Saliency detection via graph-based manifold ranking. CVPR, 3166–3173, 2013.
[8] J. Canny, A. "A Computational Approach To Edge Detection" IEEE Transactions on pattern analysis and machine intelligence, VOL. PAMI-8, NO. 6, 679 – 698, November 1986
[9] R. Achanta, A. Shaji, K. Smith, A. Lucchi, P. Fua, and S. Süsstrunk. Slic superpixels. Technical report, EPFL Technical Report no. 149300, 2010.
[10] M. Donoser, M. Urschler, M. Hirzer, and H. Bishof, “Saliency driven total variational segmentation,” in Proc. of the IEEE Int’l Conf. Computer Vision (ICCV’09), 2009.
[11] M.O.Baradez, , McGuckin, C.P., Forraz, N., Pettengell, R., Hoppe, A.: ‘Robust and automated unimodal histogram thresholding and potential applications’, Pattern Recognit., 2004, 37, (6), pp. 1131–1148
[12] A. K. Jain and R. C. Dubes. Algorithms for Clustering Data. PrenticeHall, 1988
[13] F. Kurugollu, B. Sankur, and A. Harmanci. Color image segmentation using histogram multithresholding and fusion. Image and Vison Computing, 19(13):915–928, 2001.
[14] Deng YN, Manjunath BS. Unsupervised segmentation of color-texture regions in images and video. IEEE Trans. on Pattern Analysis and Machine Intelligence, 23(8):800~810, 2001
[15] Deng YN, Kenney C, Moore MS, Manjunath BS. Peer group filtering and perceptual color image quantization. In: Cortadella J, ed. Proc. of the IEEE Int’l Symp. on Circuits and Systems. Los Alamitos, CA: IEEE Computer Society Press, 21~24. 1999