Improved CNN for the diagnosis of engine defects of 2-wheeler vehicle using wavelet synchro-squeezed transform (WSST)
作者：
Highlights：
• An improved deep learning model is developed for the diagnosis of engine defects of 2-wheeler vehicle.
• Entropy based divergence function is introduced in the existing cost function of CNN.
• The proposed function measures diversion of average sparsity from desired sparsity.
• Regularization amount of diversion is added to cost of misclassification.
• The superiority of proposed method is validated by comparing the performance with existing state-of-art works.
摘要
•An improved deep learning model is developed for the diagnosis of engine defects of 2-wheeler vehicle.•Entropy based divergence function is introduced in the existing cost function of CNN.•The proposed function measures diversion of average sparsity from desired sparsity.•Regularization amount of diversion is added to cost of misclassification.•The superiority of proposed method is validated by comparing the performance with existing state-of-art works.
论文关键词：Improved convolution neural network (CNN),Wavelet synchro-squeezed transform,Deep learning,Tacholess,2-wheeler vehicle
论文评审过程：Received 25 April 2020, Revised 4 September 2020, Accepted 7 September 2020, Available online 18 September 2020, Version of Record 22 September 2020.
论文官网地址：https://doi.org/10.1016/j.knosys.2020.106453
