A Sub-Pixel Image Registration Technique with Applications to Defect Detection

Authors: Zhen-Hui Hu, Jyh-Shong Ju, Ming-Hwei Perng

Abstract:

This paper presents a useful sub-pixel image registration method using line segments and a sub-pixel edge detector. In this approach, straight line segments are first extracted from gray images at the pixel level before applying the sub-pixel edge detector. Next, all sub-pixel line edges are mapped onto the orientation-distance parameter space to solve for line correspondence between images. Finally, the registration parameters with sub-pixel accuracy are analytically solved via two linear least-square problems. The present approach can be applied to various fields where fast registration with sub-pixel accuracy is required. To illustrate, the present approach is applied to the inspection of printed circuits on a flat panel. Numerical example shows that the present approach is effective and accurate when target images contain a sufficient number of line segments, which is true in many industrial problems.

Keywords: Defect detection, Image registration, Straight line segment, Sub-pixel.

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1058433

References:

[1] Alhichri, H. S., Kamel, M.: Multi-resolution image registration using multi-class Hausdorff fraction. Pattern Recognition Letters. 23, 279-286 (2002).
[2] Burns, J.B., Hanson, A.R.: Extracting straight lines. IEEE Trans. Pattern Anal. Mach. Intell. 4, 425-456 (1986)
[3] Canny, J.F.: A computational approach to edge detection. IEEE Trans. Pattern Anal. Mach. Intell. 8(6), 679-698 (1986)
[4] Chen, Q., Defrise, M., Deconinck, F.: Symmetric phase-only matched filtering of Fourier-Mellin transforms for image registration and recognition. IEEE Trans. Pattern Anal. Mach. Intell. 16(12), 1156-1168 (1994)
[5] Ghosal, S., Mehrotra, R.: Orthogonal moment operators for subpixel edge detection. Pattern Recognition. 26(2), 295-206 (1993).
[6] Hsieh, J.W., Liao, H.Y.M., Fan, K.C., Ko, M.T., Hung, Y.P.: Image registration using a new edge-based approach. Computer Vision and Image Understanding. 67(2), 112-130 (1997)
[7] Jain, R., Kasturi, R., Schunck B. G.: Machine Vision. McRraw-Hill (1995
[8] Keller, Y., Averbuch A., Israeli M.: Pseudopolar-based estimation of large translations, rotations, and scalings in images. IEEE Trans. Image Progress. 14(1), 12-22 (2005)
[9] Kuglin, C. D., Hines, D. C.: The phase correlation image alignment method. In: IEEE 1975 Conf. Cybernetics and Society. pp. 163-165. September (1975)
[10] Li, H. L., Chakrabarti, C.: Motion estimation of two-dimensional objects based on the straight line Hough transform: a new approach. Pattern Recognition. 29(8), 1245-1258 (1996)
[11] Nevatia, R., Babu, K.R.: Linear feature extraction and description. Comput. Graphics Image Process. 13, 257-269 (1980)
[12] Qu, Y. D., Cui, C. S., Chen, S. B., Li, J. Q.: A fast subpixel edge detection method using Sobel-Zernike moments operator. Image and Vision Computing, 23(1), 11-17 (2005).
[13] Shi, W., Shaker, A.: The line-based transformation model (LBTM) for image-to-image registration of high-resolution satellite image data. Int. J.
[14] Remote Sensing. 27(14), 3002-3023 (2006)
[15] Stockman, G. C. , Kopstein, S., Benett, S.:Matching images to models for registration and object detection via clustering. IEEE Trans. Pattern Anal. Mach. Intell. 4, 229-241 (1982)
[16] Yi, X., Camps, O. I.: Line-based recognition using a multidimensional Hausdorff distance. IEEE Trans. Pattern Anal. Mach. Intell. 21(8), 901-916 (1999)
[17] Zitovà, B., Flusser, J.: Image registration methods: a survey. Image and Vision Computing. 21, 977-1000 (2003).