TDP: Two-dimensional perceptron for image recognition
作者：
Highlights：
• A TDP has a new network architecture and an innovative computation process of hidden neurons.
• It has an independent global network structure without using the convolutional layer.
• It has local connections rather than full connections.
• It uses matrix multiplication instead of convolution operation.
• It gets direct input of two-dimensional data for further processing.
摘要
•A TDP has a new network architecture and an innovative computation process of hidden neurons.•It has an independent global network structure without using the convolutional layer.•It has local connections rather than full connections.•It uses matrix multiplication instead of convolution operation.•It gets direct input of two-dimensional data for further processing.
论文关键词：Convolutional neural network,Multilayer perceptron,Two-dimensional perceptron,Recognition performance,F1 score
论文评审过程：Received 25 November 2018, Revised 2 February 2020, Accepted 3 February 2020, Available online 8 February 2020, Version of Record 4 April 2020.
论文官网地址：https://doi.org/10.1016/j.knosys.2020.105615
