PCB matching based on SURF

被引：3

作者：

Dong, Zhi-jie ^{[1
]}

Ye, Feng ^{[1
]}

Li, Di ^{[1
]}

Huang, Jie-xian ^{[1
]}

机构：

[1] S China Univ Technol, Sch Mech & Automot Engn, Guangzhou, Guangdong, Peoples R China

来源：

CIRCUIT WORLD | 2012年 / 38卷 / 03期

关键词：

Image processing; Printed circuits; Transforms; Programming and algorithm theory; Speed-up robust feature; SURF; Hough transform; Image matching; SIFT;

D O I：

10.1108/03056121211250678

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

Purpose - The purpose of this paper is to study the application of feature-based image matching algorithm for PCB matching without using special fiducial marks. Design/methodology/approach - Speed-up robust feature (SURF) is applied to extract the interest points in PCB images. An advanced threshold is set to reject the interest points with low responses to speed up feature computation. In order to improve the performance for rotation, the descriptors are based on multi-orientations. The many-to-many tentative correspondences are determined with a maximum distance. Hough transform is used to reject the mismatches and the affine parameters are computed with a square-least solution. Findings - Results show that the method proposed in this paper can match the PCB images without using special fiducial marks effectively. The image matching algorithm shows a better performance for image rotation than the standard SURF and it succeeds in matching the image including repetitive patterns which will deteriorate the distinctiveness of feature descriptors. Research limitations/implications - Additional orientations produce more descriptors so that it takes extra time for feature description and matching. Originality/value - The paper proposes a SURF-based image matching algorithm to match the PCB images without special fiducial marks. This can reduce the complexity of PCB production. The image matching algorithm is robust to image rotation and repetitive patterns and can be used in other applications of image matching.

引用

页码：153 / 162

页数：10

共 11 条

[1]

Abdel-Hakim A. E., 2006, P IEEE COMP SOC C CO, V2, P1978, DOI DOI 10.1109/CVPR.2006.95

[2]

[Anonymous], 2009, Int. J. Image Process., DOI DOI 10.1007/S11270-006-2859-8

[3] SURF: Speeded up robust features [J].

Bay, Herbert ;

Tuytelaars, Tinne ;

Van Gool, Luc .

COMPUTER VISION - ECCV 2006 , PT 1, PROCEEDINGS, 2006, 3951 :404-417

[4]

BROWN M., 2002, BRIT MACHINE VISION, P656, DOI DOI 10.5244/C.16.23

[5] SVD-matching using SIFT features [J].

Delponte, Elisabetta ;

Isgro, Francesco ;

Odone, Francesca ;

Verri, Alessandro .

GRAPHICAL MODELS, 2006, 68 (5-6) :415-431

[6] Correction of FPC solder position error based on mutual information entropy [J].

Huang, Jie-xian ;

Ye, Feng ;

Dong, Zhi-Jie ;

Li, Di .

CIRCUIT WORLD, 2011, 37 (01) :46-51

[7]

Ke Y, 2004, PROC CVPR IEEE, P506

[8] Distinctive image features from scale-invariant keypoints [J].

Lowe, DG .

INTERNATIONAL JOURNAL OF COMPUTER VISION, 2004, 60 (02) :91-110

[9] A performance evaluation of local descriptors [J].

Mikolajczyk, K ;

Schmid, C .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2005, 27 (10) :1615-1630

[10]

Se S, 2002, 2002 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS, VOLS 1-3, PROCEEDINGS, P226, DOI 10.1109/IRDS.2002.1041393

← 1 2 →