一种基于忆阻器实现的用于图像去噪与识别的量化卷积神经网络。
A Quantized Convolutional Neural Network Implemented With Memristor for Image Denoising and Recognition.
作者信息
Zhang Yuejun, Wu Zhixin, Liu Shuzhi, Guo Zhecheng, Chen Qilai, Gao Pingqi, Wang Pengjun, Liu Gang
机构信息
Faculty of Electrical Engineering and Computer Science, Ningbo University, Ningbo, China.
Department of Micro/Nano Electronics, School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai, China.
出版信息
Front Neurosci. 2021 Sep 16;15:717222. doi: 10.3389/fnins.2021.717222. eCollection 2021.
The interference of noise will cause the degradation of image quality, which can have a negative impact on the subsequent image processing and visual effect. Although the existing image denoising algorithms are relatively perfect, their computational efficiency is restricted by the performance of the computer, and the computational process consumes a lot of energy. In this paper, we propose a method for image denoising and recognition based on multi-conductance states of memristor devices. By regulating the evolution of Pt/ZnO/Pt memristor wires, 26 continuous conductance states were obtained. The image feature preservation and noise reduction are realized the mapping between the conductance state and the image pixel. Furthermore, weight quantization of convolutional neural network is realized based on multi-conductance states. The simulation results show the feasibility of CNN for image denoising and recognition based on multi-conductance states. This method has a certain guiding significance for the construction of high-performance image noise reduction hardware system.
噪声干扰会导致图像质量下降，这会对后续的图像处理和视觉效果产生负面影响。尽管现有的图像去噪算法相对完善，但其计算效率受到计算机性能的限制，并且计算过程消耗大量能量。在本文中，我们提出了一种基于忆阻器器件多电导状态的图像去噪与识别方法。通过调节Pt/ZnO/Pt忆阻线的演化，获得了26个连续的电导状态。通过电导状态与图像像素之间的映射实现了图像特征保留和降噪。此外，基于多电导状态实现了卷积神经网络的权重量化。仿真结果表明了基于多电导状态的卷积神经网络用于图像去噪与识别的可行性。该方法对高性能图像降噪硬件系统的构建具有一定的指导意义。
Zotero 插件