欠采样
像素
计算机科学
人工智能
图像质量
稳健性(进化)
小型化
卷积神经网络
光学
计算机视觉
材料科学
图像(数学)
物理
基因
纳米技术
化学
生物化学
作者
Ikuo Hoshi,Tomoyoshi Shimobaba,Takashi Kakue,Tomoyoshi Ito
出处
期刊:Optics Express
[Optica Publishing Group]
日期:2020-10-13
卷期号:28 (23): 34069-34069
被引量:51
摘要
Single-pixel imaging allows for high-speed imaging, miniaturization of optical systems, and imaging over a broad wavelength range, which is difficult by conventional imaging sensors, such as pixel arrays. However, a challenge in single-pixel imaging is low image quality in the presence of undersampling. Deep learning is an effective method for solving this challenge; however, a large amount of memory is required for the internal parameters. In this study, we propose single-pixel imaging based on a recurrent neural network. The proposed approach succeeds in reducing the internal parameters, reconstructing images with higher quality, and showing robustness to noise.
科研通智能强力驱动
Strongly Powered by AbleSci AI