Algorithms for subpixel registration
Computer Vision, Graphics, and Image Processing
A Rule Based Approach for Visual Pattern Inspection
IEEE Transactions on Pattern Analysis and Machine Intelligence
Automated visual inspection: 1981 to 1987
Computer Vision, Graphics, and Image Processing
Self-organization and associative memory: 3rd edition
Self-organization and associative memory: 3rd edition
A survey of automated visual inspection
Computer Vision and Image Understanding
Automatic PCB inspection algorithms: a survey
Computer Vision and Image Understanding
An accurate and fast pattern localization algorithm for automated visual inspection
Real-Time Imaging - Special issue on real-time defect detection
Neural Networks: A Comprehensive Foundation
Neural Networks: A Comprehensive Foundation
Numerical Methods for Unconstrained Optimization and Nonlinear Equations (Classics in Applied Mathematics, 16)
An Efficient Algorithm for Detecting Faces from Color Images
PCM '02 Proceedings of the Third IEEE Pacific Rim Conference on Multimedia: Advances in Multimedia Information Processing
Pattern Recognition Letters
Hi-index | 0.00 |
The need for accurate and efficient pattern localization prevails in many industrial applications, such as automated visual inspection and factory automation. The image reference approach is very popular in automatic visual inspection due to its general application to a variety of inspection tasks. However, it requires precise alignment of the inspection pattern in the image. To achieve precise pattern alignment, traditional template matching is extremely time-consuming when the search space is large. In this paper, we present a new hybrid image alignment system for fast and accurate pattern localization in a large search space. This pattern localization algorithm is very useful for applications in automated visual inspection and pick-and-place applications. The proposed hybrid alignment algorithm comprises a hierarchical nearest-neighbor search process and an optical-flow based energy minimization method. The hierarchical nearest-neighbor search, which is based on the learning-from-examples principle, produces rough estimates of the transformation parameters for the initial guesses of the optical-flow based energy minimization method. The energy minimization process iteratively refines the estimation results and provides associated confidence measures. Experimental results are shown to demonstrate the accuracy and efficiency of the proposed algorithm.