采用几何复杂度的室外场景图像分割和深度生成

任艳楠, 刘琚, 元辉, 顾凌晨

任艳楠, 刘琚, 元辉, 顾凌晨. 采用几何复杂度的室外场景图像分割和深度生成[J]. 信号处理, 2018, 34(5): 531-538. DOI: 10.16798/j.issn.1003-0530.2018.05.004
引用本文: 任艳楠, 刘琚, 元辉, 顾凌晨. 采用几何复杂度的室外场景图像分割和深度生成[J]. 信号处理, 2018, 34(5): 531-538. DOI: 10.16798/j.issn.1003-0530.2018.05.004
REN Yan-nan, LIU Ju, YUAN Hui, GU Ling-chen. Outdoor Image Segmentation and Depth Generation Based on Geometry Complexity[J]. JOURNAL OF SIGNAL PROCESSING, 2018, 34(5): 531-538. DOI: 10.16798/j.issn.1003-0530.2018.05.004
Citation: REN Yan-nan, LIU Ju, YUAN Hui, GU Ling-chen. Outdoor Image Segmentation and Depth Generation Based on Geometry Complexity[J]. JOURNAL OF SIGNAL PROCESSING, 2018, 34(5): 531-538. DOI: 10.16798/j.issn.1003-0530.2018.05.004

采用几何复杂度的室外场景图像分割和深度生成

基金项目: 国家自然科学基金(61571274);山东省重点研发计划(2017CXGC1504,2015ZDXX0801A01);山东大学青年学者计划(2015WLJH39);山东省自然科学杰出青年科学基金(JQ201614);山东省自然科学基金(ZR2014FM012)
详细信息
  • 中图分类号: TN911.73

Outdoor Image Segmentation and Depth Generation Based on Geometry Complexity

  • 摘要: 本文提出一种采用几何复杂度的室外场景图像几何分割和深度生成算法。该算法首先通过图像中主要线段的角度统计分布将室外场景图像的几何结构规划为四种类型;然后,利用meanshift分割算法将输入图像分割成若干小区域,依据该图像的场景几何结构将这些小的区域逐步融合成为三个大的区域,每个区域具有一致的深度分布特点,由此实现输入图像的几何分割;最后,根据几何类型定义标准的深度图,结合输入图像的几何分割结果获得图像的深度图。实验结果表明可以通过简单的线段角度统计分布实现图像的几何分割,并进一步获得图像的深度图,与已有算法相比,提出的算法可以更好地保持深度图细节,更接近场景的真实的深度信息。
    Abstract: The paper presents an image segmentation and depth generation algorithm based on geometry complexity which is applied to outdoor scenes. Firstly, the angle statistical distribution of main lines in the input image is calculated and then the outdoor scenes are classified into four geometric categories. Secondly, the input image is divided into many small regions using the mean-shift segmentation algorithm, and then these regions are merged into three big regions based on the scene geometry category results. Each big region is in coherent depth distribution. Thirdly, a depth map can be generated based on the geometry segmented result and its standard depth map. Experimental results shows that proposed method can obtain an effective image geometric segmentation and meanwhile get a depth map with better details and more close to the true depth information of the scene.
  • [1] C. Fehn.Depth-image-based rendering (DIBR), compression and transmission for a new approach on 3d-tv[J] [J].Proc. of SPIE, 2004, 5291:93-104
    [2] Ge Guo, Nan Zhang, Longshe Huo, et al.2d to 3d conversion based on edge defocus and segmentation [C] . IEEE International Conf. on Acoustics, Speech and Signal Processing (ICASSP), 2008 : 2181-2184.
    [3] Kaiming He, Jian Sun, Xiaoou Tang.Single Image Haze Removal Using Dark Channel Prior[C]. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2009: 1956 – 1963.
    [4] Guillem Palou, Philippe Salembier.Occlusion-based depth ordering on monocular images with binary partition tree[C]. IEEE International Conf. on Acoustics, Speech and Signal Processing (ICASSP), 2011:1993 – 1996.
    [5] Guillem Palou, Philippe Salembier.From local occlusion cues to global monocular depth estimation[C]. IEEE International Conf. on Acoustics, Speech and Signal Processing (ICASSP), 2012: 793-796.
    [6] Ashutosh Saxena, Min Sun, Andrew Y.Ng. Make3D: learning 3D scene structure from a singlestill image[J].IEEE trans. on Pattern Analysis and Machine Intelligence, 2009, 31(5):824-840
    [7] H.Mohaghegh, N. Karimi, S. M. R. Soroushmehr, et al. Single image depth estimation using joint local-global features[C]. International Conference on Pattern Recognition (ICPR), 2016:27-732.
    [8] M.Mancini,GCostante,P. Valigi,et al. Toward domain independence for learning-based monocular depth estimation[J].IEEE Robotics and Automation Letters, 2017, 2(3):1778-1785
    [9] A.Martínez-Usó,PLatorre-Carmona,J. M. Sotoca,et al. Depth estimation in integral imaging based on a maximum voting strategy[J].Journal of Display Technology, 2016, 12(12):1715-1723
    [10] D.Hoiem, A.A. Efros, M. Hebert. Geometric context from a single image[C]. IEEE International Conf. on Computer Vision (ICCV), 2005:654-661.
    [11] Nedovic V., Smeulders A.W.M., Redert A., et al. Depth information by stage classification[C]. IEEE International Conf. on Computer Vision (ICCV), 2007:1-8.
    [12] C.Jung,CKim. Real-time estimation of 3D scene geometry from a single image[J].Pattern Recognition, 2012, 45(9):3256-3269
    [13] Hyukzae Lee, Chanho Jung, Changick Kim.Depth map estimation based on geometric scene categorization[C]. The 19th Korea -Japan Joint Workshop on Frontiers of Computer Vision, 2013.
    [14] Rafael C., Richard E. Woods. Digital Image Processing [M]. 3rd ed. Prentice Hall, Upper Saddle River, 2008.
    [15] D.Comaniciu,PMeer. Mean shift: a robust approach toward feature space analysis[J].IEEE trans. on Pattern Analysis and Machine Intelligence, 2002, 24(5):603-619
    [16] L.Vincent,PSoille. Watersheds in digital spaces: an efficient algorithm based on immersion simulations[J].IEEE trans on Pattern Analysis and Machine Intelligence, 1991, 13(6):583-598
    [17] B.Sumengen. Variational image segmentation and curve evolution on natural images [D] . University of California.
    [18] Jifeng Ning, Lei Zhang, David Zhang, et al.Interactive image segmentation by maximal similarity based region merging[J]. [J].pattern recognition, 2010, 43:445-456
  • 期刊类型引用(1)

    1. 曾军英,冯武林,秦传波,甘俊英,翟懿奎,王璠,朱伯远. 实时自适应的立体匹配网络算法. 信号处理. 2019(05): 843-849 . 本站查看

    其他类型引用(2)

计量
  • 文章访问数:  78
  • HTML全文浏览量:  6
  • PDF下载量:  865
  • 被引次数: 3
出版历程
  • 收稿日期:  2017-10-25
  • 修回日期:  2018-02-26
  • 发布日期:  2018-05-24

目录

    /

    返回文章
    返回