Digital image processing (2nd ed.)
Digital image processing (2nd ed.)
Texture Measures for Carpet Wear Assessment
IEEE Transactions on Pattern Analysis and Machine Intelligence
Automated inspection of printed circuit boards through machine vision
Computers in Industry
Techniques and standards for image, video, and audio coding
Techniques and standards for image, video, and audio coding
Graphical Models and Image Processing
Model-based discontinuity evaluation in the DCT domain
Signal Processing
Robust Defect Segmentation in Woven Fabrics
CVPR '98 Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
IEEE Transactions on Computers
Fast algorithm for computing discrete cosine transform
IEEE Transactions on Signal Processing
A refined fast 2-D discrete cosine transform algorithm
IEEE Transactions on Signal Processing
Variable temporal-length 3-D discrete cosine transform coding
IEEE Transactions on Image Processing
Wavelet-based principal component analysis applied to automated surface defect detection
ACC'08 Proceedings of the WSEAS International Conference on Applied Computing Conference
WSEAS Transactions on Computer Research
Expert Systems with Applications: An International Journal
Hi-index | 0.00 |
Passive components, owing to their low or no power consumption, are widely used in modern electronic devices. Nevertheless, tiny defects that often appear in the surface of passive components impair not only their appearances but also their functions. This paper proposes a global approach for the automated visual inspection of tiny surface defects in SBL (Surface Barrier Layer) chips, whose random surface texture contains no repetitions of basic texture primitives. The proposed method, taking advantage of the DCT decomposition and the cumulative sum techniques, does not requires textural features, the lack of which often limits the application of feature extraction-based methods. We apply the cumulative sum algorithm to the odd-odd matrix that gathers most power spectra in the decomposed DCT frequency domain, and select the large-magnitude frequency values that represent the background texture of the surface. Then, by reconstructing the frequency matrix without the selected frequency values, we eliminate random texture patterns and retain anomalies in the restored image. Experimental results demonstrate the effectiveness of the proposed method in inspecting tiny defects in random textures.