Optoelectronics Letters, Volume. 18, Issue 3, 181(2022)

Optimization of optical convolution kernel of optoelectronic hybrid convolution neural network

Xiaofeng XU1... Lianqing ZHU1,2,*, Wei ZHUANG2, Dongliang ZHANG2, Lidan LU3 and Pei YUAN3 |Show fewer author(s)
Author Affiliations
  • 1School of Electro-Optical Engineering, Changchun University of Science and Technology, Jilin 130022, China
  • 2Key Laboratory of the Ministry of Education for Optoelectronic Measurement Technology and Instrument, Beijing Information Science and Technology University, Beijing 100192, China
  • 3Beijing Laboratory of Optical Fiber Sensing and System, Beijing Information Science and Technology University, Beijing 100016, China
  • show less

    To enhance the optical computation's utilization efficiency, we develop an optimization method for optical convolution kernel in the optoelectronic hybrid convolution neural network (OHCNN). To comply with the actual calculation process, the convolution kernel is expanded from single-channel to two-channel, containing positive and negative weights. The Fashion-MNIST dataset is used to test the network architecture's accuracy, and the accuracy is improved by 7.5% with the optimized optical convolution kernel. The energy efficiency ratio (EER) of two-channel network is 46.7% higher than that of the single-channel network, and it is 2.53 times of that of traditional electronic products.

    Tools

    Get Citation

    Copy Citation Text

    XU Xiaofeng, ZHU Lianqing, ZHUANG Wei, ZHANG Dongliang, LU Lidan, YUAN Pei. Optimization of optical convolution kernel of optoelectronic hybrid convolution neural network[J]. Optoelectronics Letters, 2022, 18(3): 181

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Received: Dec. 4, 2021

    Accepted: Jan. 25, 2022

    Published Online: Jan. 20, 2023

    The Author Email: Lianqing ZHU (zhulianqing2021@163.com)

    DOI:10.1007/s11801-022-1183-x

    Topics