Control chart recognition based on the parallel model of CNN and LSTM with GA optimization
作者：
Highlights：
• A novel approach is proposed for the applications in control chart patterns classification.
• Feature fusion reduction were applied to extract and reduce the computational complexity.
• Convolutional neural network and long-short term memory were integrated as a classifier to obtain the results.
• The genetic algorithm was implemented to search for the optimal structure of the classifier.
• Simulation tests showed an accuracy of up to 99.85% in control chart patterns recognition.
摘要
•A novel approach is proposed for the applications in control chart patterns classification.•Feature fusion reduction were applied to extract and reduce the computational complexity.•Convolutional neural network and long-short term memory were integrated as a classifier to obtain the results.•The genetic algorithm was implemented to search for the optimal structure of the classifier.•Simulation tests showed an accuracy of up to 99.85% in control chart patterns recognition.
论文关键词：Control chart patterns recognition,Fusion feature reduction,Convolutional auto-encoder,Long-short term memory,Convolutional neural network,Genetic algorithm
论文评审过程：Received 4 December 2020, Revised 7 June 2021, Accepted 27 July 2021, Available online 31 July 2021, Version of Record 5 August 2021.
论文官网地址：https://doi.org/10.1016/j.eswa.2021.115689
