An improved all-optical diffractive deep neural network with less parameters for gesture recognition
An improved all-optical diffractive deep neural network with less parameters for gesture recognition
复制标题
DOI：
10.1016/j.jvcir.2022.103688
复制
发表时间：
2022-11-18
影响因子：
2.6
通讯作者：
Abdi-Ghaleh, Reza
中科院分区：
文献类型：
--
作者：
Zhou, Yuanguo;Shui, Shan;Abdi-Ghaleh, Reza
作为光学机器学习的框架，全光衍射神经网络（D2NN）在特征检测和目标分类方面提供了理想的结果，目前引起了光学和光子学界的高度兴趣。在本文中，我们将改进的D2NN架构应用于手势识别领域，其特征是比先前文献中常见的MNIST手写识别具有更复杂的轮廓。所提出的网络结构结合了类小波相位调制模式技术和基于全光神经网络的高速公路网络。通过调制入射光的相位，类小波图案可以大大减少网络层的参数。此外，还采用高速公路网络来解决训练过程中梯度消失的现象。在实验中，我们在数字上实现了识别十种不同手势的盲测准确率达到 95.6%，并且参数数量仅为常规 D2NN 的 3%。可靠性测试和分析表明，该方法是一种低参数、高效的解决方案，有望实现各种机器学习任务。
As a framework of optical machine learning, all-optical diffractive neural network (D2NN) has delivered an ideal outcome of feature detection and target classification, currently raising high interest in the optics and photonics community. In this paper, we applied an improved D2NN architecture to the field of gesture recognition, which features more complicated contour than the common MNIST handwriting recognition in the previous literature. The proposed network structure incorporates the wavelet-like phase modulation pattern technique and the highway network on the basis of all-optical neural network. Through modulating the phase of incident light, the wavelet-like pattern can substantially reduce the parameters in the network layer. In addition, a highway network is employed to address the vanishing gradient phenomenon in the training process. In the experiment, we numerically achieved blind testing accuracy of 95.6% for identifying ten different gestures, and the number of parameters is only 3% of the regular D2NN. Reliability test and analysis show that the proposed method is a high-efficiency solution with low-parameters expecting for implementation of various machine learning tasks.