Optical Panel Inspection Using Explicit Band Gaussian Filtering Methods in Discrete Cosine Domain
Abstract
:1. Introduction
2. Literature Review
3. Research Method
3.1. Two-Dimensional Frequency Transform
3.2. Analyses of Frequency Power Spectrums
3.3. Frequency Spectrum Filtering
3.3.1. Threshold Filtering (TF) Approach
3.3.2. Band Filtering (BF) Approach
3.3.3. Band-Threshold Filtering (BTF) Approach
3.3.4. Band Gaussian Filtering (BGF) Approach
3.3.5. Double-Band Gaussian Filtering (DBGF) Approach
3.4. Image Rebuild and Defect Segmentation
4. Experiments and Results
4.1. Performance Evaluation of the Explicit Filtering Methods with Various Parameter Settings
4.1.1. Bandwidths and Frequency Thresholds of Filtering in BTF Approach
4.1.2. Bandwidths and Energy Threshold Coefficients of Filtering in DBGF Approach
4.2. Comparisons of the Different Band Filtering Methods
4.3. Large-Sample Experiments
4.4. Comparison of Performance Evaluation Indexes of Different Detection Methods
4.5. Robustness Testing of Flaw Detections for Various Cutting Angles and Background Textures with the Proposed Methods
4.5.1. Performance of Using Different Band Angle Filters on Defect Detection
4.5.2. Detection of CTP Images with Different Background Textures
5. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Acknowledgments
Conflicts of Interest
References
- Ebayyeh, M.A.; Mousavi, A. A review and analysis of automatic optical inspection and quality monitoring methods in electronic industry. IEEE Access 2020, 8, 183192–183271. [Google Scholar] [CrossRef]
- Luo, Q.; Fang, X.; Liu, L.; Yang, C.; Sun, Y. Automated Visual Defect Detection for Flat Steel Surface: A Survey. IEEE Trans. Instrum. Meas. 2020, 69, 626–644. [Google Scholar] [CrossRef]
- Adamo, F.; Attivissimo, F.; Di Nisio, A.; Savino, M. A low-cost inspection system for online defects assessment in satin glass. Measurement 2009, 42, 1304–1311. [Google Scholar] [CrossRef]
- Ng, H.-F. Automatic thresholding for defect detection. Pattern Recognit. Lett. 2006, 27, 1644–1649. [Google Scholar] [CrossRef]
- Otsu, N. A threshold selection method from gray-level histograms. IEEE Trans. Syst. Man Cybern. 1979, 9, 62–66. [Google Scholar] [CrossRef]
- Li, W.-C.; Tsai, D.-M. Wavelet-based defect detection in solar wafer images with inhomogeneous texture. Pattern Recognit. 2012, 45, 742–756. [Google Scholar]
- Lin, H.-D. Tiny surface defect inspection of electronic passive components using discrete cosine transform decomposition and cumulative sum techniques. Image Vis. Comput. 2008, 26, 603–621. [Google Scholar] [CrossRef]
- Lettry, L.; Perdoch, M.; Vanhoey, K.; Van Gool, L. Repeated pattern detection using CNN activations. In Proceedings of the 2017 IEEE Winter Conference on Applications of Computer Vision (WACV), Santa Rosa, CA, USA, 24–31 March 2017; pp. 47–55. [Google Scholar]
- Lin, H.-D.; Tsai, H.-H. Automated quality inspection of surface defects on touch panels. J. Chin. Inst. Ind. Eng. 2012, 29, 291–302. [Google Scholar] [CrossRef]
- Hung, M.-H.; Hsieh, C.-H. A novel algorithm for flaw inspection of touch panels. Image Vis. Comput. 2015, 41, 11–25. [Google Scholar]
- Liang, L.Q.; Li, D.; Fu, X.; Zhang, W.J. Touch screen flaw inspection based on sparse representation in low-resolution images. Multimed. Tools Appl. 2016, 75, 2655–2666. [Google Scholar] [CrossRef]
- Chiu, Y.-S.P.; Lin, H.-D. Creation of image models for inspecting visual flaws on capacitive touch screens. J. Appl. Eng. Sci. 2018, 16, 333–342. [Google Scholar]
- Jian, C.X.; Gao, J.; Ao, Y. Automatic surface flaw detection for mobile phone screen glass based on machine vision. Appl. Soft Comput. 2018, 52, 348–358. [Google Scholar]
- Ye, R.; Pan, C.S.; Chang, M.; Yu, Q. Intelligent flaw classification system based on deep learning. Adv. Mech. Eng. 2018, 10, 1–7. [Google Scholar] [CrossRef]
- Lei, J.; Gao, X.; Feng, Z.; Qiu, H.; Song, M. Scale insensitive and focus driven mobile screen flaw detection in industry. Neurocomputing 2018, 294, 72–81. [Google Scholar] [CrossRef]
- Ye, R.; Chang, M.; Pan, C.S.; Chiang, C.A.; Gabayno, J.L. High-resolution optical inspection system for fast detection and classification of surface flaws. Int. J. Optomechatronics 2018, 21, 1–10. [Google Scholar] [CrossRef]
- Tsai, D.-M.; Hsieh, C.-Y. Automated surface inspection for directional textures. Image Vis. Comput. 1999, 18, 49–62. [Google Scholar] [CrossRef]
- Perng, D.-B.; Chen, S.-H. Automatic surface inspection for directional textures using nonnegative matrix factorization. Int. J. Adv. Manuf. Technol. 2010, 48, 671–689. [Google Scholar]
- Chen, Y.C.; Yu, J.H.; Xie, M.C.; Shiou, F.J. Automated optical inspection system for analogical resistance type touch panel. Int. J. Phys. Sci. 2011, 6, 5141–5152. [Google Scholar]
- Jiang, C.; Quan, Y.; Lin, X. Defect detection of capacitive touch panel using a nonnegative matrix factorization and tolerance model. Appl. Opt. 2016, 55, 2331–2338. [Google Scholar]
- Ahmed, N.; Natarajan, T.; Rao, K.R. Discrete cosine transform. IEEE Trans. Comput. 1974, 23, 90–93. [Google Scholar]
- Gonzalez, R.C.; Woods, R.E. Digital Image Processing, 4th ed.; Pearson: New York, NY, USA, 2018. [Google Scholar]
- Pogrebnyak, O.; Lukin, V.V. Wiener discrete cosine transform-based image filtering. J. Electron. Imaging 2012, 21, 043020. [Google Scholar] [CrossRef]
- Zhang, X.; Wu, H.; Ma, Y. A new auto-focus measure based on medium frequency discrete cosine transform filtering and discrete cosine transform. Appl. Comput. Harmon. Anal. 2016, 40, 430–437. [Google Scholar] [CrossRef]
- Zhang, J.; Liao, Y.; Zhu, X.; Wang, H.; Ding, J. A Deep Learning Approach in the Discrete Cosine Transform Domain to Median Filtering Forensics. IEEE Signal Process. Lett. 2020, 27, 276–280. [Google Scholar] [CrossRef]
- Lin, H.-D.; Jiang, J.-D. Applying discrete cosine transform and grey relational analysis to surface defect detection of LEDs. J. Chin. Inst. Ind. Eng. 2007, 24, 458–467. [Google Scholar]
Types | Pattern 1 | Pattern 2 | Pattern 3 | Pattern 4 | Pattern 5 | Pattern 6 |
---|---|---|---|---|---|---|
Texture complexity | Complex texture | Complex texture | Moderate texture | Moderate texture | Simple texture | Simple texture |
Texture figure | ||||||
Texture image |
Band Filtering Methods | (α)%: 0~60% and (1 − β)%: 70~100% | |||
---|---|---|---|---|
3W-BF | 3W-BTF | 3W-BGF | 3W-DBGF | |
Parameter settings | (= 1) | (= 1, = 150) | (= 1, = 3) | (= 2, = 3; = 2, = 3) |
AUC (%) | 94.93 | 97.46 | 97.57 | 97.94 |
Indicator | Spatial Domain | Frequency Domain | ||||
---|---|---|---|---|---|---|
Iterative [21] | Otsu [5] | Tsai and Hsieh [17] | Perng and Chen [18] | Lin and Tsai [9] | Proposed Method | |
DFT + BF | DCT + TF | DFT + MC-BF | DCT + 3W-DBGF | |||
k | -- | -- | 2.3 | 2.3 | 2.3 | 2.3 |
1 − β (%) | 99.89 | 99.94 | 76.75 | 88.78 | 92.72 | 94.21 |
α (%) | 41.84 | 44.16 | 36.68 | 3.23 | 2.98 | 1.97 |
CR (%) | 58.26 | 55.95 | 63.38 | 96.75 | 97.01 | 98.04 |
Time (s) | 0.0078 | 0.0047 | 1.03 | 0.26 | 2.96 | 1.62 |
Filtering parameter | -- | -- | W = 1 | T = 100 | = 1, = 0.5 | (= 2,= 3; = 2,= 3) |
Complex Texture (Background Texture-2) | Moderate Texture (Background Texture-3) | Simple Texture (Background Texture-5) | |
---|---|---|---|
Sample images (Normal samples) | |||
Parameter settings | (= 2, = 3; = 2, = 3) | (= 2, = 1; = 2, = 1) | (= 1, = 0.5; = 1, = 0.5) |
k | 1.3 | −0.1 | 1 |
1 − β (%) | 93.11 | 92.06 | 92.66 |
α (%) | 0.53 | 3.02 | 0.31 |
CR (%) | 99.44 | 87.73 | 99.68 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Lin, H.-D.; Tsai, H.-H.; Lin, C.-H.; Chang, H.-T. Optical Panel Inspection Using Explicit Band Gaussian Filtering Methods in Discrete Cosine Domain. Sensors 2023, 23, 1737. https://doi.org/10.3390/s23031737
Lin H-D, Tsai H-H, Lin C-H, Chang H-T. Optical Panel Inspection Using Explicit Band Gaussian Filtering Methods in Discrete Cosine Domain. Sensors. 2023; 23(3):1737. https://doi.org/10.3390/s23031737
Chicago/Turabian StyleLin, Hong-Dar, Huan-Hua Tsai, Chou-Hsien Lin, and Hung-Tso Chang. 2023. "Optical Panel Inspection Using Explicit Band Gaussian Filtering Methods in Discrete Cosine Domain" Sensors 23, no. 3: 1737. https://doi.org/10.3390/s23031737