[1]游达章１,２,周宏耀１,等.基于特征优化的 Census立体匹配方法[J].湖北工业大学学报,2024,39(1):41-45.
　YOU Dazhang,ZHOU Hongyao,ZHANG Yepeng.Research and Implementation of Census Stereo Matching Method Based on Feature Information Optimization[J].,2024,39(1):41-45.
点击复制

基于特征优化的 Census立体匹配方法()
分享到：

《湖北工业大学学报》[ISSN:1003-4684/CN:42-1752/Z]

卷:
39
期数:
2024年第1期
页码:
41-45
栏目:
出版日期:
2024-02-20

文章信息/Info

Title:
Research and Implementation of Census Stereo Matching Method Based on Feature Information Optimization
文章编号:
1003-4684(2023)05-0041-05
作者:
游达章１; ２; 周宏耀１; ２; 张业鹏１; ２
(１ 湖北工业大学机械工程学院,湖北 武汉 ４３００６８;２ 湖北省现代制造质量工程重点实验室,湖北 武汉 ４３００６８)
Author(s):
YOU Dazhang1; 2;  ZHOU Hongyao 1; 2;  ZHANG Yepeng1; 2
（1 School of Mechanical Engineering, Hubei Univ. of Tech., Wuhan 430068, China;2 Hubei Key Lab of Manufacture Quality Engineering, Wuhan 430068, China)
关键词:
立体匹配; Census; 特征信息优化; 降采样策略
Keywords:
stereo matching;  Census;  characteristic information optimization;  downsampling strategy
分类号:
TP３９１.４１
文献标志码:
A
摘要:
针对传统 Census立体匹配算法在弱纹理和边缘区域匹配精度较差的问题,提出一种基于特征信息优化的代价计算方法,在窗口中融入更多的差异信息以获得更精确的像素视差值.随后采用多方向路径独立的线扫描优化计算聚合代价以进一步提高匹配精度.为获得更好的遮挡区域匹配效果,提出一种基于差异填充的视差优化方法,对遮挡像素进行识别和视差填充.为提高算法的效率,提出一种基于降采样策略的算法运行模式,通过缩小视差搜索范围以减少硬件负荷.最后以五组标准图像为输入进行改进 Census算法性能检验,结果显示,平均误匹配率为６．１２％,较改进前降低了２．４５％,算法效率平均提升１７．７％.
Abstract:
Aiming at the problem of poor matching accuracy of traditional Census stereo matching algorithm in weak texture and edge areas, we propose a cost calculation method based on feature information optimization, which integrates more difference information into the window to obtain more accurate pixel disparity value. Subsequently, multidirectional path independent line scan optimization was used to calculate the aggregation cost to further improve the matching accuracy. In order to obtain better occlusion area matching effect, a disparity optimization method based on difference filling is proposed to identify the occlusion pixels and make disparity filling. In order to improve the efficiency of the algorithm, a new algorithm operation mode based on the downsampling strategy is proposed to reduce the hardware load by narrowing the disparity search range. Finally, the performance test of the improved Census algorithm was conducted with five sets of standard images as input. The results showed that the average mismatching rate was 6.12%, which was 2.45% lower than before the improvement, and the average efficiency of the algorithm increased by 17.7%.

参考文献/References:

[１]　LEE J,JUN DAEYOON,etal．Improved censustransformfornoiserobuststereomatching[J]．OpticalEngineering,２０１６,５５(０６):０６３１０７Ｇ１Ｇ０６３１０７Ｇ１０．[２]　HAMZAH,ROSTAM AFFENDI,IBRAHIM HAIＧDI,etal．Stereomatchingalgorithmbasedonperpixeldifferenceadjustment,iterativeguidedfilterandgraphsegmentation[J]．Journalofvisualcommunication &imagerepresentation,２０１７,４２(Jan．):１４５Ｇ１６０．[３]　蒋文萍,汪凌阳,韩文超,等．基于改进 Census变换的自适应 局 部 立 体 匹 配 [J]．电 子 测 量 技 术,２０２２,４５(１３):８２Ｇ８７．[４]　虞文杰,叶嵩,郭毓,等．基于改进 Census变换与多特征融合 的 立 体 匹 配 算 法 [J]．激 光 与 光 电 子 学 进 展,２０２２,５９(０８):１５７Ｇ１６３．[５]　曹毅,包祥威,吴翔．基于重排序 Census变换的半全局立体匹配算法[J]．电子测量技术,２０２１,４４(２４):４０Ｇ４６．[６]　余嘉昕,王春媛,韩华,等．基于融合代价和优化引导滤波的立体 匹 配 算 法 [J]．计 算 机 工 程,２０２３,４９(０３):２５７Ｇ２６２,２７０．[７]　KARIMI HOSSEIN, DOLATI ARDESHIR,LAYEGHIKAMRAN．Stereo matching by usinganewlocaldescriptorbasedonsignandorderinbeliefpropagationalgorithm[J]．Displays:TechnologyandApplications,２０２２,７５(１０):１７３１Ｇ１７３９．[８]　MUKHERJEE,SUBHAYAN,GUDDETI,etal．Ahybridalgorithmfordisparitycalculationfromsparsedisparityestimatesbasedonstereovision[C]．∥２０１４４４ 湖　北　工　业　大　学　学　报 ２０２４年第１期　International Conference on Signal Processing andCommunications:２０１４ International Conference onSignalProcessingandCommunications(SPCOM),２２Ｇ２５July２０１４,Bangalore,India．:InstituteofElectricalandElectronicsEngineers,２０１４:１Ｇ６．[９]　IRIJANTIE,NAYAN M Y,YUSOFF M Z．Localstereo matching algorithm using smallＧcolorcensusandsparseadaptivesupportweight[C]．∥２０１１ NaＧtionalPostgraduateConference．[v．２]．:IEEE,２０１１:９２９Ｇ９３３．[１０]HIRSCHMULLER H．AccurateandefficientstereoprocessingbysemiＧglobalmatchingand mutualinforＧmation[C]．ComputerVisionandPatternRecognition,２００５．CVPR２００５．IEEEComputerSocietyConferenceonvol．２．２００５:８０７Ｇ８１４．[１１]唐笑虎,胡丹,刘凯．一种极线近似的双目结构光相位立体匹配 方 法 [J/OL]．强 激 光 与 粒 子 束:１Ｇ６[２０２２Ｇ０９Ｇ２８]．http:∥kns．cnki．net/kcms/detail/５１．１３１１．O４．２０２２０７２２．１９１６．００２．html．[１２]李涵,黄妙华．自适应区域划分立体匹配算法[J/OL]．激光与光电子学进展:１Ｇ１７[２０２２Ｇ０９Ｇ２８]．http:∥kns．cnki．net/kcms/detail/３１．１６９０．TN．２０２２０７１４．１８３０．４７３．html．[１３]ZHUORAN L X．MultileveldisparityreconstructionnetworkforrealＧtimestereomatching[J]．上海交通大学学报(英文版),２０２２,２７(０５):７１５Ｇ７２２．[１４]陶洋,田家旺．匹配代价融合与分层迭代优化的半全局立体匹配[J/OL]．小型微型计算机系统:１Ｇ８[２０２２Ｇ０９Ｇ２８]．http:∥kns．cnki．net/kcms/detail/２１．１１０６．TP．２０２２０７０７．１５０５．００６．html．[１５]杨记鑫,胡伟霞,赵杰．基于双目视觉的标定和立体匹配的研究[J]．电子设计工程,２０２２,３０(１３):５０Ｇ５３．[１６]XIN M,ZHICHENGZ,DANFENG W,etal．AdapＧtivedeconvolutionＧbasedstereo matchingnetforlocalstereomatching[J]．AppliedSciences,２０２２,１２(０４):２０８６．[１７]王正家,陈长乐,徐研彦,等．基于跨尺度PatchMatch的立体匹配算法[J]．电子测量技术,２０２２,４５(１２):１１４Ｇ１１９．[１８]JIYANGQ,LIANGL．Thestereomatchingalgorithmbasedonanimprovedadaptivesupportwindow[J]．IETImageProcessing,２０２２,１６(１０):２８０３Ｇ２８１６．[１９]余嘉昕,王春媛,韩华,等．基于融合代价和优化引导滤波的立体 匹 配 算 法 [J]．计 算 机 工 程,２０２３,４９(０３):２５７Ｇ２６２,２７０．[２０]KARNAUKHOV V N,KOBER VI,MOZEROV MG,etal．Robuststereo matchingusingphasefeaturesbasedonthewalshＧhadamardtransform[J]．JournalofCommunicationsTechnologyandElectronics,２０２１,６６(１２):１４３８Ｇ１４４３．

备注/Memo

备注/Memo:
[收稿日期]２０２２ -１１ -１４[基金项目]国家自然科学基金(５１８７５１８０)[第一作者]游达章(１９７５－),男,湖北武汉人,工学博士,湖北工业大学教授,研究方向为机器人与智能控制、数控技术、故障预测与可靠性技术.[通信作者]周宏耀(１９９７－),男,湖北麻城人,湖北工业大学硕士研究生,研究方向为立体匹配.
更新日期/Last Update: 2024-03-14