Variational Bayesian Subgroup Adaptive Sparse Component Extraction for Diagnostic Imaging System

Bin Gao, Peng Lu, Wai Lok Woo, Gui Yun Tian, Yuyu Zhu, Martin Johnston
2018 IEEE transactions on industrial electronics (1982. Print)  
A novel unsupervised sparse component extraction algorithm is proposed for detecting micro defects when employing a thermography imaging system. The proposed approach is developed using the Variational Bayesian framework. This enables a fully automated determination of the model parameters and bypasses the need for human intervention in manually selecting the appropriate image contrast frames. An internal sub-sparse grouping mechanism and adaptive fine-tuning strategy have been built to control
more » ... the sparsity of the solution. The proposed algorithm is computationally affordable and yields a high accuracy objective performance. Experimental tests on both artificial and natural defects have been conducted to verify the efficacy of the proposed method. Index Terms -Low-rank decomposition, Variational Bayesian (VB), diagnostic imaging system, sparse decomposition.
doi:10.1109/tie.2018.2801809 fatcat:mz3ljl2qhzbgfkuoavbhk4qvjm