In this study, an local invariant generalized Houghtransform (LI-GHT) method is proposed for integrated circuit (IC) visual positioning. The original generalized Hough transform (GHT) is robust to external noise; however, it is not suitable for visual positioning of IC chips due to the four-dimensionality (4D) of parameter space which leads to the substantial storage requirement and high computational complexity. The proposed LI-GHT method can reduce the dimensionality of parameter space to 2D thanks to the rotational invariance of local invariant geometric feature and it can estimate the accuracy position and rotation angle of IC chips in real-time under noise and blur influence. The experiment results show that the proposed LI-GHT can estimate position and rotation angle of IC chips with high accuracy and fast speed. The proposed LI-GHT algorithm was implemented in IC visual positioning system of radio frequency identification (RFID) packaging equipment.
A Local Invariant Generalized Hough Transform Method for Integrated Circuit Visual Positioning
2013-11-04
oai:zenodo.org:1089529
Article (Journal)
Electronic Resource
English
DDC: | 629 |
Linear generalized Hough transform and its parallelization
British Library Online Contents | 1993
|Generalized Hough Transform for Arabic Optical Character Recognition
British Library Conference Proceedings | 2003
|British Library Online Contents | 1993
|Visual guidance control based on the Hough transform
IEEE | 2000
|