Spatial Filtering with Multi-scale Segmentation Based on Gaussian Function

Authors:
Chi-Fan Chen;Chia-Hsin Liang
Affiliations:
Gallant Precision Machining Co., Ltd, R&D Div., Visual Dept., Hsinchu, Taiwan 308;Gallant Precision Machining Co., Ltd, R&D Div., Visual Dept., Hsinchu, Taiwan 308
Venue:
ISVC '08 Proceedings of the 4th International Symposium on Advances in Visual Computing, Part II
Year:
2008

Citing 5
Cited 0

A Theory for Multiresolution Signal Decomposition: The Wavelet Representation

IEEE Transactions on Pattern Analysis and Machine Intelligence
A Multiresolution Hierarchical Approach to Image Segmentation Based on Intensity Extrema

IEEE Transactions on Pattern Analysis and Machine Intelligence
A multiscale wavelet solver with O(n) complexity

Journal of Computational Physics
Digital Image Processing Using MATLAB

Digital Image Processing Using MATLAB
Defect detection in periodically patterned surfaces using independent component analysis

Pattern Recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

This research mainly focuses in developing the algorithm for vision inspection inside the TFT-LCD panels. The purpose of this algorithm is to extract the defects precisely from the source images with regular patterns and the falloff of illumination caused by the vignetting effect. Specific spatial masks are generated from the Gaussian function by means of the concept of normal distribution. The source image is decomposed by the Gaussian masks into different spatial frequencies for the scale-space representation. The experimental result shows that the defect image can be effectively segmented from the source image with the smallest size less than 0.5pixel. The optimized algorithm is also provided in this paper for shortening the process time to meet the industrial needs in AOI equipment. This paper provides a novel and complete solution for scale manipulation in image processing.