Abstract

Genetically modified food has always been a hot issue in the field of food safety. In order to realize the detection of genetically modified materials, a bionic recognition model of a multi-weight vector neural network is proposed by combining a multi-weight vector neural network with terahertz time domain spectroscopy. In this paper, for each class of samples, 50 samples are randomly selected as the training set, a multiweight vector neural network bionic recognition model is established, and 50 samples are selected as the first test set to verify the recognition rate. Other dissimilar samples are used as the second test set to verify their misjudgment rate. The experimental results show that the model can effectively identify transgenic materials with similar spectral characteristics. The model proposed in this paper provides a new method for the detection and identification of genetically modified organisms.

© 2021 Optical Society of America

Sweep frequency method with variance weight probability for temperature extraction of the Brillouin gain spectrum based on an artificial neural network

Ming Hai Wang, Yang Sui, Wei Nan Zhou, Wei Dong, and Xin Dong Zhang
Opt. Express 29(18) 28994-29006 (2021)

Multi-wavelength diffractive neural network with the weighting method

Jianan Feng, Hang Chen, Dahai Yang, Junbo Hao, Jie Lin, and Peng Jin
Opt. Express 31(20) 33113-33122 (2023)

Delay-weight plasticity-based supervised learning in optical spiking neural networks

Yanan Han, Shuiying Xiang, Zhenxing Ren, Chentao Fu, Aijun Wen, and Yue Hao
Photon. Res. 9(4) B119-B127 (2021)

Multichannel left-subtract-right feature vector piston error detection method based on a convolutional neural network

Peng-Fei Wang, Hui Zhao, Xiao-Peng Xie, Ya-Ting Zhang, Chuang Li, and Xue-Wu Fan
Opt. Express 29(14) 21320-21335 (2021)

Optical neural network using vector-feature extraction

Yasunori Kuratomi, Akio Takimoto, Koji Akiyama, and Hisahito Ogawa
Appl. Opt. 32(29) 5750-5758 (1993)