基于图割的低景深图像自动分割

刘毅; 陈圣磊; 冯国富; 黄兵; 夏德深

doi:10.16383/j.aas.2015.c140734

基于图割的低景深图像自动分割

doi: 10.16383/j.aas.2015.c140734

1.
南京审计学院工学院南京 210094;
2.
南京理工大学计算机科学与技术学院南京 210029

基金项目:

国家自然科学基金(61473157), 江苏省高校自然科学研究项目(13KJ B520013, 14KJB520019)资助

详细信息

作者简介:
陈圣磊博士,南京审计学院副教授,澳大利亚莫纳什大学信息技术学院兼职研究员.主要研究方向为数据挖掘与机器学习.E-mail:tristan_chen@126.com

计量
- 文章访问数: 1653
- HTML全文浏览量: 115
- PDF下载量: 2043
- 被引次数: 0
出版历程
- 收稿日期: 2014-10-22
- 修回日期: 2015-04-11
- 刊出日期: 2015-08-20

Automatic Segmentation of Images with Low Depth of Field Based on Graph Cuts

1.
School of Technology, Nanjing Audit University, Nanjing 210094;
2.
School of Computer Science and Technology, Nanjing University of Science and Technology, Nanjing 210029

Funds:

Supported by National Natural Science Foundation of China (61473157) and Natural Science Foundation of the Jiangsu Higher Education Institutions of China (13KJB520013, 14KJB5 20019)

摘要

摘要: 结合图割算法,提出了一种针对低景深(Depth of field, DOF)图像的自动分割模型.首先,通过改进的点锐度算法得到图像的点锐度图, 并结合图像的颜色特征,得到一个四维的特征向量.其次, 通过对图像点锐度图强边缘的计算,利用图像清晰部分边缘较连续, 模糊部分边缘较弱、连续性较差的特点得到图像初步的前景/背景区域. 然后,对前景/背景的颜色和点锐度特征进行高斯混合模型(Gaussian mixture model, GMM)建模,结合全局、局部自适应的λ值,对图割算法的Shrinking bias 现象进行改善.最后,通过迭代的图割算法对前景/背景区域进行修正. 实验结果表明,该模型鲁棒性较高,分割结果更加精确.
- 图割 /
- 低景深 /
- 点锐度图 /
- 高斯混合模型
Abstract: An automatic segmentation model combined with graph cuts algorithm for low depth of field (DOF) images is proposed. Firstly, the point sharpness algorithm is improved to extract the point sharpness map of the image. In combination with color features, a four dimensional vector is constructed. Secondly, strong edges of the point sharpness map are exacted and the characteristics that the edges of clear part of an image are commonly continuous and the edges of blurred part are weak and discontinuous are used to get the preliminary foreground/background regions. Then, Gaussian mixture model (GMM) is used to model the features of point sharpness and color and by using global and local adaptive λ the shrinking bias problem of graph cuts algorithm is improved effectively. Finally, the iterative graph cuts algorithm is used to revise the foreground/background regions. Experiments show that the proposed segmentation model is more robust and more accurate.
- Graph cuts /
- low depth of field /
- point sharpness map /
- Gaussian mixture model (GMM)

HTML全文

参考文献(20)

[1]	Kim C. Segmenting a low-depth-of-field image using morphological filters and region merging. IEEE Transactions on Image Processing, 2005, 14(10): 1503-1511
[2]	Deng Xiao-Ling, Ni Jiang-Qun, Li Zhen, Dai Fen. Foreground extraction from low depth-of-field images based on colour-texture and HOS features. Acta Automatica Sinica, 2013, 39(6): 846-851(邓小玲, 倪江群, 李震, 代芬. 多特征融合的低景深图像前景提取算法. 自动化学报, 2013, 39(6): 846-851)
[3]	Li H L, Ngan K N. Unsupervized video segmentation with low depth of field. IEEE Transactions on Circuits and Systems for Video Technology, 2007, 17(12): 1742-1751
[4]	Li H L, Ngan K N. Learning to extract focused objects from low DOF images. IEEE Transactions on Circuits and Systems for Video Technology, 2011, 21(11): 1571-1580
[5]	Graf F, Kriegel H P, Weiler M. Robust segmentation of relevant regions in low depth of field images. In: Proceedings of the 18th International Conference on Image Processing. Brussels, Belgium: IEEE, 2011. 2861-2864
[6]	Chen T T, Li H L. Segmenting focused objects based on the amplitude decomposition model. Pattern Recognition Letters, 2012, 33(12): 1536-1542
[7]	Konik H, Neverova N. Edge-based method for sharp region extraction from low depth of field images. In: Proceedings of the 2002 International Conference on Visual Communications and Image Processing. San Diego, USA: IEEE, 2012. 1-6
[8]	Mei J Y, Si Y L, Gao H J. A curve evolution approach for unsupervised segmentation of images with low depth of field. IEEE Transactions on Image Processing, 2013, 22(10): 4086 -4095
[9]	Boykov Y Y, Jolly M P. Interactive graph cuts for optimal boundary and region segmentation of objects in N-D images. In: Proceedings of the 2001 International Conference on Computer Vision. Vancouver, Canada: IEEE, 2001. 105 -112
[10]	Liu Song-Tao, Yin Fu-Liang. The basic principle and its new advances of image segmentation methods based on graph cuts. Acta Automatica Sinica, 2012, 38(6): 911-922(刘松涛, 殷福亮. 基于图割的图像分割方法及其新进展. 自动化学报, 2012, 38(6): 911-922)
[11]	Liu Song-Tao, Wang Hui-Li, Yin Fu-Liang. Interactive ship infrared image segmentation method based on graph cut and fuzzy connectedness. Acta Automatica Sinica, 2012, 38(11): 1735-1750(刘松涛, 王慧丽, 殷福亮. 基于图割和模糊连接度的交互式舰船红外图像分割方法. 自动化学报, 2012, 38(11): 1735-1750)
[12]	Zhou H L, Zheng J M, Wei L. Texture aware image segmentation using graph cuts and active contours. Pattern Recognition, 2013, 46(6): 1719-1733
[13]	Zhang Shi-Hui, Luo Yan-Qing, Kong Ling-Fu. Shadow detection based on graph cuts for a single image. Acta Automatica Sinica, 2014, 40(10): 2306-2315(张世辉, 罗艳青, 孔令富. 基于图割的单幅图像影子检测. 自动化学报, 2014, 40(10): 2306-2315)
[14]	Wang Hong-Nan, Zhong Wen, Wang Jing, Xia De-Shen. Research of measurement for digital image definition. Journal of Image and Graphics, 2004, 9(7): 828-831(王鸿南, 钟文, 汪静, 夏德深. 基图像清晰度评价方法研究. 中国图象图形学报, 2004, 9(7): 828-831)
[15]	Marziliano P, Dufaux F, Winkler S, Ebrahimi T. Perceptual blur and ringing metrics: application to JPEG2000. Signal Processing: Image Communication, 2004, 19(2): 163-172
[16]	Pratt W K. Digital Image Processing. New York: John Wiley and Sons, Inc., 1978. 514
[17]	Rother C, Kolmogorov V, Blake A. GrabCut: interactive foreground extraction using iterated graph cuts. In: Proceedings of the 31st ACM International Conference on Computer Graphics and Interactive Techniques. Los Angeles, USA: ACM, 2004. 309-314
[18]	Candemir S, Akgül Y S. Adaptive regularization parameter for graph cut segmentation. In: Proceedings of the 7th International Conference on Image Analysis and Recognition. Póvoa de Varzim, Portugal: Springer, 2010. 117-126
[19]	Candemir S, Akgül Y S. Statistical significance based graph cut segmentation for shrinking bias. In: Proceedings of the 8th International Conference on Image Analysis and Recognition. Burnaby, Canada: Springer, 2011. 304-313
[20]	Goldberger J, Gordon S, Greenspan H. An efficient image similarity measure based on approximations of KL-divergence between two Gaussian mixtures. In: Proceedings of the 10th International Conference on Computer Vision and Pattern Recognition. Nice, France: IEEE, 2003. 487-493