An Improved GMM-Based Algorithm With Optimal Multi-Color Subspaces for Color Difference Classification of Solar Cells

Automatic color classification for solar cells is challenging because of the tiny color difference and low contrast. To address this problem, a color feature selection and classification frame is proposed in this paper. First, an intuitive multi-color space feature performance evaluation scheme is presented to select the optimal color subspaces that help to enormously enlarge the tiny color difference of solar cell images. And the optimal color subspaces can also be illustrated by employing multi-color space visualization method with combinational Mosaic images. Second, a nine-dimensional feature vector consisting of mean, variance, and skewness in the three optimal subspaces is extracted by utilizing the serial fusion technique. Third, an improved Gaussian mixture model in supervised manner for color classification is proposed by employing a k-means method based on adjacent rules, which helps to eliminate isolated points and enhances the classification performance. Finally, experimental results show that the overall performance of the proposed method achieves 97.9%, and outperform other algorithms especially for the tiny color difference of solar cell images.

[1]  Stefania Matteoli,et al.  A Spectroscopy-Based Approach for Automated Nondestructive Maturity Grading of Peach Fruits , 2015, IEEE Sensors Journal.

[2]  Monson H. Hayes,et al.  Adaptive defogging with color correction in the HSV color space for consumer surveillance system , 2012, IEEE Transactions on Consumer Electronics.

[3]  Xi Chen,et al.  Manifold Regularized Gaussian Mixture Model for Semi-supervised Clustering , 2013, 2013 2nd IAPR Asian Conference on Pattern Recognition.

[4]  H. Hajj,et al.  Wafer Classification Using Support Vector Machines , 2012, IEEE Transactions on Semiconductor Manufacturing.

[5]  Roberlanio Oliveira Melo,et al.  Leak Detection of Natural Gas with Base on the Components of Color Spaces RGB and HSI Using Novelty Filter , 2014, IEEE Latin America Transactions.

[6]  Sandro Cumani,et al.  Feature Fusion for Fingerprint Liveness Detection: a Comparative Study , 2017, IEEE Access.

[7]  Guo Lili,et al.  The comparison of optimizing SVM by GA and grid search , 2017, 2017 13th IEEE International Conference on Electronic Measurement & Instruments (ICEMI).

[8]  Kok-Meng Lee,et al.  Effects of Classification Methods on Color-Based Feature Detection With Food Processing Applications , 2007, IEEE Transactions on Automation Science and Engineering.

[9]  Zhen Zhou,et al.  Application of Multi-color Space Feature Fusion in Color Difference Processing , 2014, 2014 Fourth International Conference on Instrumentation and Measurement, Computer, Communication and Control.

[10]  K.P. White,et al.  Classification of Defect Clusters on Semiconductor Wafers Via the Hough Transformation , 2008, IEEE Transactions on Semiconductor Manufacturing.

[11]  Wenyong Lin.  An improved GMM-based clustering algorithm for efficient speaker identification , 2015, 2015 4th International Conference on Computer Science and Network Technology (ICCSNT).

[12]  King Ngi Ngan,et al.  Blind Image Quality Assessment Based on Multichannel Feature Fusion and Label Transfer , 2016, IEEE Transactions on Circuits and Systems for Video Technology.

[13]  Markus A. Stricker,et al.  Similarity of color images , 1995, Electronic Imaging.

[14]  Takeshi Nakazawa,et al.  Wafer Map Defect Pattern Classification and Image Retrieval Using Convolutional Neural Network , 2018, IEEE Transactions on Semiconductor Manufacturing.

[15]  Yuan Liu,et al.  Integrated color defect detection method for polysilicon wafers using machine vision , 2014 .

[16]  Kadim Tasdemir,et al.  Classification of black mold contaminated figs by hyperspectral imaging , 2015, 2015 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT).

[17]  Bipan Tudu,et al.  A Machine Vision Technique for Grading of Harvested Mangoes Based on Maturity and Quality , 2016, IEEE Sensors Journal.

[18]  Yu Wang,et al.  Credible Intervals for Precision and Recall Based on a K-Fold Cross-Validated Beta Distribution , 2016, Neural Computation.

[19]  Dimitris E. Koulouriotis,et al.  A Unified Methodology for Computing Accurate Quaternion Color Moments and Moment Invariants , 2014, IEEE Transactions on Image Processing.

[20]  Dah-Jye Lee,et al.  Rapid Color Grading for Fruit Quality Evaluation Using Direct Color Mapping , 2011, IEEE Transactions on Automation Science and Engineering.

[21]  Paul Scheunders,et al.  Multisource Classification of Color and Hyperspectral Images Using Color Attribute Profiles and Composite Decision Fusion , 2012, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[22]  Ghalia Tello,et al.  Deep-Structured Machine Learning Model for the Recognition of Mixed-Defect Patterns in Semiconductor Fabrication Processes , 2018, IEEE Transactions on Semiconductor Manufacturing.

[23]  Gerald Schaefer,et al.  Thermography based breast cancer analysis using statistical features and fuzzy classification , 2009, Pattern Recognit..