An Optimization Algorithm of Robust Principal Component Analysis and Its Application

2019 2nd International Conference on Advanced Materials, Intelligent Manufacturing and Automation

Xia, Xinyuan^1 ; Gao, Fei^1

Beijing Institute of Technology, Beijing, China^1

关键词: Application effect;    Approximate zeros;    Comparative analysis;    Inertial momentum;    Model parameters;    Objective optimization;    Optimization algorithms;    Robust principal component analysis;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/569/5/052099/pdf
DOI  :  10.1088/1757-899X/569/5/052099

来源: IOP

【 摘 要 】

With the rapid development of robust principal component analysis (RPCA), it has been widely used in signal processing, pattern recognition, computer vision and other fields. The RPCA model has characteristics of complete reconstruction of the original signal from the noise pollution, high-dimensional and high-order complex signals. In order to solve the problems of slow iteration speed and low recovery accuracy in typical algorithms, an improved robust principal component analysis (RPCA) algorithm is studied. Firstly, the idea of smooth approximate zero norm is introduced to build the objective optimization function, then the inertial momentum is used to optimize each iteration of the matrix recovery process, finally, the model parameters are optimized by the grid method. Through simulation and comparative analysis, the results show that the improved algorithm has high accuracy, fast processing speed and remarkable application effect in field of logging data processing.