计算机科学 ›› 2021, Vol. 48 ›› Issue (11A): 314-318.doi: 10.11896/jsjkx.201200264

• 图像处理& 多媒体技术 • 上一篇    下一篇

基于图像分割的自适应窗口双目立体匹配算法研究

曹林, 于威威   

  1. 上海海事大学信息工程学院 上海201306
  • 出版日期:2021-11-10 发布日期:2021-11-12
  • 通讯作者: 于威威(wwyu@shmtu.edu.cn)
  • 作者简介:lcao768@163.com

Adaptive Window Binocular Stereo Matching Algorithm Based on Image Segmentation

CAO Lin, YU Wei-wei   

  1. School of Information Engineering,Shanghai University of Maritime,Shanghai 201306,China
  • Online:2021-11-10 Published:2021-11-12
  • About author:CAO Lin,born in 1991,postgraduate.Her main research interests include computer vision and image processing.
    YU Wei-wei,born in 1978,Ph.D,asso-ciate professor.Her main research interests include pattern recognition,image processing and data mining.

摘要: 针对传统双目立体匹配算法采用固定窗口导致弱纹理区域匹配精度较低的问题,提出了一种基于图像分割的自适应窗口立体匹配算法。首先,采用Mean-shift算法对图像进行分割,之后对分割图像进行局部子区灰度标准差统计,在此基础上提出了一种根据纹理丰富程度进行窗口大小自适应设定的算子。基于自适应窗口大小设定,组合使用Census变换和梯度值计算匹配代价,并分别通过自适应权重代价聚合及“胜者为王”策略进行初始视差计算,最后利用左右视差一致性原则和加权中值滤波得到稠密视差图。采用提出的自适应窗口匹配算法与固定窗口匹配算法对Middlebury数据集上的标准图片进行匹配实验,实验结果表明,所提算法的平均匹配错误率为2.04%,相比对比算法,所提方法的匹配错误率分别降低了4.5%和7.9%。

关键词: 立体匹配, 图像分割, 弱纹理, 自适应窗口, 自适应权重

Abstract: Aiming at the problem that the traditional binocular stereo matching algorithm uses fixed window,which leads to low matching accuracy in weak texture regions,an adaptive window stereo matching algorithm based on image segmentation is proposed.Firstly,the mean shift algorithm is used to segment the image,and then the gray standard deviation of local sub regions is calculated.Based on this,an adaptive window size setting operator is proposed according to the texture richness.Based on the adaptive window size setting,the matching cost is calculated by combining census transform and gradient value,and the initial disparity is calculated by adaptive weight cost aggregation and “winner takeall” strategy respectively.Finally,the dense disparity map is obtained by using the principle of left and right disparity consistency and weighted median filtering.The adaptive window matching algorithm and fixed window matching algorithm proposed in this paper are used to match standard images on Middlebury dataset.The experimental results show that the average matching error rate of the proposed algorithm is 2.04%,which is 4.5% and 7.9% lower than that of the contrast algorithm.

Key words: Stereo matching, Image segmentation, Weak texture, Adaptive window, Adaptive weight

中图分类号: 

  • TP311.5
[1]ZHANG S.Recent progresses on real-time 3D shape measurement using digital fringe projection techniques[J].Optics and Lasers in Engineering,2009,48(2):149-158.
[2]KIEU H,PAN T Y,WANG Z Y,et al.Accurate 3D shapemeasurement of multiple separate objects with stereo vision[J].Measurement Science and Technology,2014,25(3):1-7.
[3]JIANG S,HONG Z,ZHANG Y,et al.Automatic path planning and navigation with stereo cameras[C]//2014 3rd International Workshop on Earth Observation and Remote Sensing Applications (EORSA).IEEE,2014:289-293.
[4]SUHR J K.Automatic free parking space detection by using motion stereo-based 3D reconstruction[J].Machine Vision &Applications,2010,21(2):163-176.
[5]BRUNO F,BIANCO G,MUZZUPAPPA M,et al.Experimentation of structured light and stereo vision for underwater 3D reconstruction[J].ISPRS Journal of Photogrammetry and Remote Sensing,2011,66(4):508-518.
[6]ZHANG J,MCMILLAN L,YU J.Robust Tracking and Stereo Matching under Variable Illumination[C]//IEEE Computer Society Conference on Computer Vision & Pattern Recognition.2006(1):871-878.
[7]TIPPETTS B,LEE D J,LILLYWHITE K,et al.Review of ste-reo vision algorithms and their suitability for resource-limited systems[J].Journal of Real-Time Image Processing,2016,11(1):5-25.
[8]BROWN M Z,BURSCHKA D,HAGER G D.Advances in Computational Stereo[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2003,25(8):993-1008.
[9]KLAUS A.Segment-based stereo matching using belief propagation and a self-adapting dissimilarity measure[J].InternationalConference on Pattern Recognition,2006(3):15-18
[10]HU T,QI B,WU T,et al.Stereo matching using weighted dynamic programming on a single-direction four-connected tree[J].Computer Vision & Image Understanding,2012,116(8):908-921.
[11]XU Y Y,XU X Y,YU R.Disparity Optimization Algorithm for Stereo Matching Using Improved Guided Filter[J].Journal of Advanced Computational Intelligence and Intelligent Informati-cs,2019,23(4):625-633.
[12]YU H B,HU Y L,XU J.Stereo matching algorithm based on multi feature fusion and tree structure cost aggregation[J].Journal of Shanghai University (Natural Science Edition),2019,25(1):66-74.
[13]ZHANG Z.A flexible New Technique for Camera Calibration[J].IEEETransactions on Pattern and Machine Intelligence,2000,22(11):1330-1334.
[14]TAO H,SAWHNEY H S,KUMAR R.Aglobal matchingframework for stereo computation[C]//IEEE International Conference on Computer Vision Vancouver.2001:523-539.
[15]ZHANG Y F,LI X F,TIAN X D.Stereo matching algorithm based on image segmentation[J].Computer Applications,2020,40(5):1415-1420.
[16]CMANICIU D,MEER P.Mean Shift:A Robust Approach Toward Feature Space Analysis[J].IEEE TransactionsPattern Analysis and Machine Intelligence,2002,24(5):603-619.
[17]YANG Q,WANG L,YANG R,et al.Stereo Matching with Color-Weighted Correlation,Hierarchical Belief Propagation,and Occlusion Handling[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2009,31(3):492-504.
[18]SCHARSTEIN D,H HIRSCHMÜLLER,KITAJIMA Y,et al.High-Resolution Stereo Datasets with Subpixel-Accurate Ground Truth[C]//German Conference on Pattern Recognition.Springer International Publishing,2014:31-42.
[19]HIRSCHMÜLLER H,SCHARSTEIN D.Evaluation of CostFunctions for Stereo Matching[C]//IEEE Conference on Com-puter Vision & Pattern Recognition.IEEE,2007:1-8.
[1] 张鹏, 王新晴, 肖毅, 段宝国, 许鸿辉. 基于语义边缘驱动的实时双目深度估计算法[J]. 计算机科学, 2021, 48(9): 216-222.
[2] 叶中玉, 吴梦麟. 融合时序监督和注意力机制的脉络膜新生血管分割[J]. 计算机科学, 2021, 48(8): 118-124.
[3] 金海燕, 彭晶, 周挺, 肖照林. 基于Graph Cuts多特征选择的双目图像分割方法[J]. 计算机科学, 2021, 48(8): 150-156.
[4] 许华杰, 张晨强, 苏国韶. 基于深层卷积残差网络的航拍图建筑物精确分割方法[J]. 计算机科学, 2021, 48(8): 169-174.
[5] 杨秀璋, 武帅, 夏换, 于小民. 基于自适应图像增强技术的水族文字提取与识别研究[J]. 计算机科学, 2021, 48(6A): 74-79.
[6] 顾兴健, 朱剑峰, 任守纲, 熊迎军, 徐焕良. 多尺度U网络实现番茄叶部病斑分割与识别[J]. 计算机科学, 2021, 48(11A): 360-366.
[7] 桑苗苗, 彭进先, 达通航, 张旭峰. 基于PatchMatch的半全局高效双目立体匹配算法[J]. 计算机科学, 2021, 48(1): 204-208.
[8] 朱玲莹, 桑庆兵, 顾婷婷. 基于视差信息的无参考立体图像质量评价[J]. 计算机科学, 2020, 47(9): 150-156.
[9] 程中建, 周双娥, 李康. 基于多尺度自适应权重的稀疏表示目标跟踪算法[J]. 计算机科学, 2020, 47(6A): 181-186.
[10] 杨志伟, 戴铭, 周智恒. 基于直方图差异的工业产品表面缺陷检测方法[J]. 计算机科学, 2020, 47(6A): 247-249.
[11] 曹义亲, 段也钰, 武丹. 基于WFSOA的2D-Otsu钢轨缺陷图像分割方法[J]. 计算机科学, 2020, 47(5): 154-160.
[12] 杨婷, 罗飞, 丁炜超, 卢海峰. 一种自适应优化松弛量的装箱算法[J]. 计算机科学, 2020, 47(4): 211-216.
[13] 饶梦,苗夺谦,罗晟. 一种粗糙不确定的图像分割方法[J]. 计算机科学, 2020, 47(2): 72-75.
[14] 雷涛,连倩,加小红,刘鹏. 基于快速SLIC的图像超像素算法[J]. 计算机科学, 2020, 47(2): 143-149.
[15] 周岳勇,程江华,刘通,王洋,陈明辉. 高分辨率SAR图像道路提取综述[J]. 计算机科学, 2020, 47(1): 124-135.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
[1] 冯芙蓉, 张兆功. 目标轮廓检测技术新进展[J]. 计算机科学, 2021, 48(6A): 1 -9 .
[2] 孙正, 张小雪. 生物光声成像中声反射伪影抑制方法的研究进展[J]. 计算机科学, 2021, 48(6A): 10 -14 .
[3] 周欣, 刘硕迪, 潘薇, 陈媛媛. 自然交通场景中的车辆颜色识别[J]. 计算机科学, 2021, 48(6A): 15 -20 .
[4] 黄雪冰, 魏佳艺, 沈文宇, 凌力. 基于自适应加权重复值滤波和同态滤波的MR图像增强[J]. 计算机科学, 2021, 48(6A): 21 -27 .
[5] 江妍, 马瑜, 梁远哲, 王原, 李光昊, 马鼎. 基于分数阶麻雀搜索优化OTSU肺组织分割算法[J]. 计算机科学, 2021, 48(6A): 28 -32 .
[6] 朝乐门, 王锐. 数据科学平台:特征、技术及趋势[J]. 计算机科学, 2021, 48(8): 1 -12 .
[7] 冯霞, 胡志毅, 刘才华. 跨模态检索研究进展综述[J]. 计算机科学, 2021, 48(8): 13 -23 .
[8] 周文辉, 石敏, 朱登明, 周军. 基于残差注意力网络的地震数据超分辨率方法[J]. 计算机科学, 2021, 48(8): 24 -31 .
[9] 罗长银, 陈学斌, 马春地, 张淑芬. 基于层析分析改进的联邦平均算法[J]. 计算机科学, 2021, 48(8): 32 -40 .
[10] 朝乐门, 尹显龙. 人工智能治理理论及系统的现状与趋势[J]. 计算机科学, 2021, 48(9): 1 -8 .