Adaptive Threshold Epipolar Distance Transformation Method with Applications to Stereo Matching

doi:10.7652/xjtuxb202107015

您当前的位置：

首页 >

文章列表页 >

Adaptive Threshold Epipolar Distance Transformation Method with Applications to Stereo Matching

更新时间：2025-07-09

- Adaptive Threshold Epipolar Distance Transformation Method with Applications to Stereo Matching
- Vol. 55, Issue 7, Pages: 132-140(2021)
- 作者机构：
  
  1. 西安交通大学机械结构强度与振动国家重点实验室,西安,710049
  2. 西安交通大学陕西省先进飞行器服役环境与控制重点实验室,西安,710049
  3. 西安交通大学航天航空学院,西安,710049
- 作者简介：
- 基金信息：
- DOI：10.7652/xjtuxb202107015
  CLC： TP302.7
- Online First：10 July 2021，
  
  Published：2021
- 稿件说明：
移动端阅览
Adaptive Threshold Epipolar Distance Transformation Method with Applications to Stereo Matching[J]. 2021, 55(7): 132-140.
DOI：

Adaptive Threshold Epipolar Distance Transformation Method with Applications to Stereo Matching[J]. 2021, 55(7): 132-140. DOI： 10.7652/xjtuxb202107015.

摘要

为了提升图像在不重叠区域和弱纹理区域的匹配率

提出一种基于图像局部纹理信息的阈值自适应的极线距离变换(AT-EDT)算法。选取左右视图对应视差范围内的点

将对应点在特定窗口中的灰度信息作为参考

去除图像不重叠区域; 针对在极线距离变换(EDT)中使用固定阈值会产生二次弱纹理和重复纹理的现象

将局部标准差与灰度梯度进行融合来表示图像局部纹理信息

并由此信息获得图像中每个像素对应的距离变换阈值; 利用AT-EDT算法对图像进行变换

并通过立体匹配得到最终视差图。实验结果表明:利用AT-EDT算法可有效去除图像的不重叠区域并显著提升弱纹理区域的像素可分性

同时避免了EDT算法导致的二次弱纹理和重复纹理现象; 对变换后的图像进行立体匹配

可有效提升视差图的深度连续性; 与EDT后的图像立体匹配得到的视差图相比

AT-EDT的匹配率至少提升了20%。

Abstract

To improve the matching rate of images in non-overlapping regions and weak texture regions

an adaptive threshold epipolar distance transformation(AT-EDT)method related to image local texture information is proposed. The points in the disparity range corresponding to the left and right views are selected

and the gray information of the corresponding point in a specific window is taken as a reference to remove the non-overlapping area of the image. Aiming at the secondary weak texture and repeated texture that will appear as a fixed threshold is used in epipolar distance transformation(EDT)

the local standard deviation and gray gradient are fused to represent the local texture information of the image

and the corresponding distance transformation threshold of each pixel in the image is obtained from this information. The AT-EDT method is chosen to transform the image

and the disparity map is obtained by stereo matching. Experimental results show that the AT-EDT method can effectively remove the non-overlapping regions of the image and significantly improve the pixel separability of the weak texture regions while the secondary weak texture and repetitive texture caused by the EDT algorithm are avoided. The stereo matching of the transformed image can improve the depth continuity of disparity map. Compared with the disparity map obtained by the stereo matching of the image after EDT

the disparity map matching rate is heightened by at least 20%.

关键词

Keywords

references

BALTER M L, CHEN A I, MAGUIRE T J, et al. Adaptive kinematic control of a robotic venipuncture device based on stereo vision, ultrasound, and force guidance [J]. IEEE Transactions on Industrial Electronics, 2017, 64(2): 1626-1635.

VICENTE P, JAMONE L, BERNARDINO A. Robotic hand pose estimation based on stereo vision and GPU-enabled internal graphical simulation [J]. Journal of Intelligent Robotic Systems, 2016, 83(3/4): 339-358.

LU Bo, CHU H K, HUANG Kaicheng, et al. Surgical suture thread detection and 3-D reconstruction using a model-free approach in a calibrated stereo visual system [J]. IEEE/ASME Transactions on Mechatronics, 2020, 25(2): 792-803.

张俊勇, 伍世虔, 陈彬, 等. 基于双目视觉的零件多尺寸在线测量系统 [J]. 仪表技术与传感器, 2018(10): 75-80.

ZHANG Junyong, WU Shiqian, CHEN Bin, et al. Binocular vision based multi-dimensions on-line measuring system for workpieces [J]. Instrument Technique and Sensor, 2018(10): 75-80.

李佳宽, 孙春生, 胡艺铭, 等. 一种基于ORB特征的水下立体匹配算法 [J]. 光电工程, 2019, 46(4): 180456.

LI Jiakuan, SUN Chunsheng, HU Yiming, et al. An underwater stereo matching method based on ORB features [J]. Opto-Electronic Engineering, 2019, 46(4): 180456.

HALLEK M, BOUKAMCHA H, SMACH F, et al. Real time stereo matching using two step zero-mean SAD and dynamic programing [C]∥Proceedings of the 2018 15th International Multi-Conference on Systems, Signals Devices(SSD). Piscataway, NJ, USA: IEEE, 2018: 1234-1240.

MA Ning, MEN Yubo, MEN Chaoguang, et al. Segmentation-based stereo matching using combinatorial similarity measurement and adaptive support region [J]. Optik, 2017, 137: 124-134.

蒙雁琦, 胡改玲, 温琳鹏, 等. 区域分割的亚像素相位立体匹配算法 [J]. 西安交通大学学报, 2017, 51(10): 33-38.

MENG Yanqi, HU Gailing, WEN Linpeng, et al. A stereo matching algorithm of sub-pixel phases based on region segmention [J]. Journal of Xi'an Jiaotong University, 2017, 51(10): 33-38.

CHAI Yu, CAO Xiaojing. Stereo matching algorithm based on joint matching cost and adaptive window [C]∥Proceedings of the 2018 IEEE 3rd Advanced Information Technology, Electronic and Automation Control Conference(IAEAC). Piscataway, NJ, USA: IEEE, 2018: 442-446.

张一飞, 李新福, 田学东. 基于图像分割的立体匹配算法 [J]. 计算机应用, 2020, 40(5): 1415-1420.

ZHANG Yifei, LI Xinfu, TIAN Xuedong. Stereo matching algorithm based on image segmentation [J]. Journal of Computer Applications, 2020, 40(5): 1415-1420.

HAMZAH R A, WEI M G Y, ANWAR N S N. Development of stereo matching algorithm based on sum of absolute RGB color differences and gradient matching [J]. International Journal of Electrical and Computer Engineering(IJECE), 2020, 10(3): 2375.

SHI Hua, ZHU Hong, WANG Jing, et al. Segment-based adaptive window and multi-feature fusion for stereo matching [J]. Journal of Algorithms Computational Technology, 2016, 10(1): 3-11.

YANG Qingxiong, AHUJA N. Stereo matching using epipolar distance transform [J]. IEEE Transactions on Image Processing, 2012, 21(10): 4410-4419.

岳颀. 基于自适应权重极线距离变换的立体匹配 [J]. 科学技术与工程, 2015, 15(28): 39-42, 66.

YUE Qi. A local stereo matching algorithm based on adaptive support-weight epipolar distance transformation [J]. Science Technology and Engineering, 2015, 15(28): 39-42, 66.

符立梅, 彭国华. 基于自适应极线距离变换的立体匹配 [J]. 计算机辅助设计与图形学学报, 2015, 27(3): 394-399.

FU Limei, PENG Guohua. Adaptive epipolar distance transform-based stereo matching [J]. Journal of Computer-Aided Design Computer Graphics, 2015, 27(3): 394-399.

曹晓倩, 马彩文. 基于弱纹理检测及视差图融合的立体匹配 [J]. 仪器仪表学报, 2013, 34(9): 2074-2079.

CAO Xiaoqian, MA Caiwen. Stereo matching based on low-texture detection and disparity map fusion [J]. Chinese Journal of Scientific Instrument, 2013, 34(9): 2074-2079.

ZHANG Ke, LU Jiangbo, LAFRUIT G. Cross-based local stereo matching using orthogonal integral images [J]. IEEE Transactions on Circuits and Systems for Video Technology, 2009, 19(7): 1073-1079.

郭川. 三维重建中序列图像的立体匹配与拼接方法研究 [D]. 南昌: 南昌航空大学, 2015: 36-38.

张芬, 彭直兴. 肇事车辆刹车痕迹图像特征匹配技术仿真 [J]. 计算机仿真, 2015, 32(5): 201-204.

ZHANG Fen, PENG Zhixing. The vehicle braking track simulation image feature matching technology [J]. Computer Simulation, 2015, 32(5): 201-204.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Sensitivity Analysis of Foil Geometrical Parameters for the Third-Generation CO2 Gas Foil Bearings Based on Design of Experiments Method

Numerical Simulation of Heat-Moisture Barrier Performance of a Horizontal Air Curtain in Cold Storage

Turbulent Kinetic Energy Dissipation Mechanism of Supercritical Carbon Dioxide Leakage Flow in a Scallop Bionic Seal Cavity

Integrated and Joint Design Optimization of Centrifugal Compressor Impeller and Casing Treatment at Multiple Working Points

Numerical Simulation of the Energy Conversion Process in Thermoacoustic Engines with Different Stack Materials

Related Author

No data

Related Institution

No data

⁰