Qu Yuan, Lian Hengyu, Ding Chunxu, Liu Huazhen, Liu Linxian, Yang Jiamiao
Opt Lett. 2023 Oct 1;48(19):5025-5028. doi: 10.1364/OL.498712.
The existing implementations of reconfigurable diffractive neural networks rely on both a liquid-crystal spatial light modulator and a digital micromirror device, which results in complexity in the alignment of the optical system and a constrained computational speed. Here, we propose a superpixel diffractive neural network that leverages solely a digital micromirror device to control the neuron bias and connection. This approach considerably simplifies the optical system and achieves a computational speed of 326 Hz per neural layer. We validate our method through experiments in digit classification, achieving an accuracy of 82.6%, and action recognition, attaining a perfect accuracy of 100%. Our findings demonstrate the effectiveness of the superpixel diffractive neural network in simplifying the optical system and enhancing computational speed, opening up new possibilities for real-time optical information processing applications.
可重构衍射神经网络的现有实现方式依赖于液晶空间光调制器和数字微镜器件，这导致光学系统对准复杂且计算速度受限。在此，我们提出一种超像素衍射神经网络，该网络仅利用数字微镜器件来控制神经元偏差和连接。这种方法极大地简化了光学系统，并实现了每个神经层326赫兹的计算速度。我们通过数字分类实验验证了我们的方法，准确率达到82.6%，在动作识别方面，准确率达到了100%。我们的研究结果证明了超像素衍射神经网络在简化光学系统和提高计算速度方面的有效性，为实时光学信息处理应用开辟了新的可能性。
