|Table of Contents|

Automatic matching, stitching and fusion method of concrete bridge crack images(PDF)

长安大学学报（自然科学版）[ISSN:1006-6977/CN:61-1281/TN]

Issue:
2022年6期
Page:
33-41
Research Field:
桥梁工程·交通基础设施智能化运维技术专栏
Publishing date:

Info

Title:
Automatic matching, stitching and fusion method of concrete bridge crack images
Author(s):
YU Li-cun1; 2;  HE Shuan-hai1;  JIANG Shu-qing3;  XIANG Shui-ying3
(1. School of Highway, Chang’an University, Xi’an 710064, Shaanxi, China; 2. CCCC First Highway Consultants Co., Ltd, Xi’an 710075, Shaanxi, China; 3. State Key Laboratory of Integrated Service Networks, Xidian University, Xi’an 710071, Shaanxi, China)
Keywords:
bridge engineering;  crack stitching;  moving direct linear transformation(Moving DLT);  concrete crack;  image fusion;  visualization
PACS:
U446.3
DOI:
10.19721/j.cnki.1671-8879.2022.06.004
Abstract:
Aimed at the concrete bridge crack images collected by automatic image acquisition equipment faces several serious problems including huge amount, low matching accuracy, poor stitching and fusion effect, an end-to-end solution for concrete bridge crack from matching, stitching, fusion to merging statistics was proposed. First, in the stage of crack matching, the conventional method of feature point matching was replaced by the combination of feature point matching and search set. Then, the moving direct linear transformation(Moving DLT)and singular value decomposition(SVD)was employed to calculate the local homography matrix and implement the mapping, which solved the problem that the global homography matrix was not enough to fit all the feature points. In addition, the Alpha weighted fusion processing algorithm based on distance transformation was adopted to realize the natural fusion of stitching lines. Based on the crack image data set collected in real engineering, the stitching and fusion effects of the moving direct linear transformation algorithm and the traditional OpenCV method were comprehensively compared and analyzed from the aspects of the alignment of crack stitching lines, the accuracy and integrity of crack visualization results, and the texture consistency of non-crack regions. Besides, in the image stitching stage, the problem of repeated visualization of cracks in the overlapping region of stitching images was solved by controlling the color transformation range of cracks in source image and target image. In the stage of image fusion, mask visualization method was used to solve the problem that the color display of cracks caused by image fusion become lighter or even disappeared. Finally, according to the relative relationship between detection boxes and crack vertices, the same crack was merged and reconnected in the stitching image. The results show that the improved algorithm can effectively improve the matching accuracy and achieve good stitching and fusion effect, and can realize the rapid and seamless stitching of crack images collected by intelligent detection equipment.7 figs, 26 refs.

References:

[1] LIU Z Q,CAO Y W,WANG Y Z,et al.Computer vision-based concrete crack detection using U-net fully convolutional networks[J].Automation in Construction,2019,104:129-139.
[2]ZHANG C B,CHANG C C,JAMSHIDI M.Concrete bridge surface damage detection using a single-stage detector[J].Computer-Aided Civil and Infrastructure Engineering,2020,35(4):389-409.
[3]LI G,LIU Q W,ZHAO S M,et al.Automatic crack recognition for concrete bridges using a fully convolutional neural network and naive Bayes data fusion based on a visual detection system[J].Measurement Science and Technology,2020,31(7):075403.
[4]王 超,贾 贺,张社荣,等.基于图像的混凝土表面裂缝量化高效识别方法[J].水力发电学报,2021,40(3):134-144.
WANG Chao,JIA He,ZHANG She-rong,et al.Image-based quantitative and efficient identification method for concrete surface cracks[J].Journal of Hydroelectric Engineering,2021,40(3):134-144.
[5]SZELISKI R.Image alignment and stitching:A tutorial[J].Foundations and Trends in Computer Graphics and Vision,2007,2(1):1-104.
[6]XIE R P,YAO J,LIU K,et al.Automatic multi-image stitching for concrete bridge inspection by combining point and line features[J].Automation in Construction,2018,90:265-280.
[7]邵进达,杨 帅,程 琳.改进SIFT算法结合两级特征匹配的无人机图像匹配算法[J].计算机科学,2019,46(6):316-321.
SHAO Jin-da,YANG Shuai,CHENG Lin.UAV image matching algorithm based on improved SIFT algorithm and two-stage feature matching[J].Computer Science,2019,46(6):316-321.
[8]王 云.中小型混凝土桥梁裂缝图像检测系统研究[D].杭州:杭州电子科技大学,2020.
WANG Yun.Research on crack image detection system of small and medium-sized concrete bridge[D].Hangzhou:Hangzhou Dianzi University,2020.
[9]梅 智.基于无人机图像的混凝土坝表观裂缝监测研究[D].大连:大连理工大学,2020.
MEI Zhi.Research on the monitoring of concrete dam surface crack based on UAV image[D].Dalian:Dalian University of Technology,2020.
[10]BROWN M,LOWE D G.Recognising panoramas[C]//ICCV.Proceedings Ninth IEEE International Conference on Computer Vision.Nice:IEEE,2003:1218-1225.
[11]TUYTELAARS T,MIKOLAJCZY K.Local invariant feature detectors:A survey[J].Foundations and trends in computer graphics and vision,2008,3(3):177-280.
[12]ZITOVá B,FLUSSER J.Image registration methods:A survey[J].Image and Vision Computing,2003,21(11):977-1000.
[13]CHANDRATRE R,A CHAKKARWAR V.Image stitching using Harris and RANSAC[J].International Journal of Computer Applications,2014,89(15):14-19.
[14]ALOMRAN M,CHAI D.Feature-based panoramic image stitching[C]//IEEE.Proceedings of the 14th International Conference on Control,Automation,Robotics and Vision(ICARCV).New York:IEEE,2016:1-6.
[15]WU L,LIN X,CHEN Z,et al.Surface crack detection based on image stitching and transfer learning with pretrained convolutional neural network[J].Structural Control and Health Monitoring,2021,28(8):e2766.
[16]WON J,PARK J W,SHIM C,et al.Bridge-surface panoramic-image generation for automated bridge-inspection using deepmatching[J].Structural Health Monitoring,2020:147592172093038.
[17]YU L C,HE S H,LIU X S,et al.Intelligent crack detection and quantification in the concrete bridge:A deep learning-assisted image processing approach[J].Advances in Civil Engineering,2022,2022:1-15.
[18]ABEYWICKRAMA T,CHEEMA M A,TANIAR D.K-nearest neighbors on road networks[J].Proceedings of the VLDB Endowment,2016,9(6):492-503.
[19]曲天伟,安 波,陈桂兰.改进的RANSAC算法在图像配准中的应用[J].计算机应用,2010,30(7):1849-1851,1872.
QU Tian-wei,AN Bo,CHEN Gui-lan.Application of improved RANSAC algorithm to image registration[J].Journal of Computer Applications,2010,30(7):1849-1851,1872.
[20]BROWN M,LOWE D G.Automatic panoramic image stitching using invariant features[J].International Journal of Computer Vision,2007,74(1):59-73.
[21]ABDI H.Singular value decomposition(SVD)and generalized singular value decomposition[J].Encyclopedia of Measurement and Statistics,2007(1):907-912.
[22]SZELISKI R.Computer vision:Algorithms and applications[M].Washington DC:Springer Science & Business Media,2010.
[23]FELZENSZWALB P F,HUTTENLOCHER D.Distance transforms of sampled functions[J].Theory of computing,2012,8(1):415-428.
[24]KAEHLER A,BRADSKI G.Learning OpenCV 3:Computer vision in C++ with the OpenCV library[M].Cambridge:O’Reilly Media Inc.,2017.
[25]任 重,唐 垚,谢永斌,等.基于多频段融合的全景图像拼接技术[J].科技创新与应用,2021(1):59-61,64.
REN Chong,TANG Yao,XIE Yong-bin,et al.Panoramic image stitching technology based on multi-band fusion[J].Technology Innovation and Application,2021(1):59-61,64.
[26]ZARAGOZA J,CHIN T J,BROWN M S,et al.As-projective-as-possible image stitching with moving DLT[J].Pattern Analysis and Machine Intelligence,2014,36(7):1285-1298.

Memo

Memo:
-
Last Update: 2022-12-20