计算机科学
鬼影成像
像素
探测器
人工智能
计算机视觉
动态成像
光谱成像
图像传感器
高动态范围成像
光学
图像处理
物理
动态范围
图像(数学)
高动态范围
电信
数字图像处理
作者
Wenjie Jiang,Yongkai Yin,Junpeng Jiao,Xian Zhao,Baoqing Sun
出处
期刊:Photonics Research
[Optica Publishing Group]
日期:2022-07-19
卷期号:10 (9): 2157-2157
被引量:50
摘要
Single-pixel imaging (SPI) can capture 2D images of the target with only a nonpixelated detector, showing promising application potential in nonvisible spectral imaging, low-photon imaging, lidar, and other extreme imaging fields. However, the imaging mechanism of traditional SPI makes it difficult to achieve high imaging speed, which is a primary barrier for its widespread application. To address this issue, in this work, we propose and demonstrate a novel high-speed 2D and 3D imaging scheme based on traditional SPI, termed time-resolved single-pixel imaging (TRSPI). Previous SPI works mainly utilize correlation between a stable target and iterative illumination masks to reconstruct a single image. In TRSPI, by further exploiting correlation information between a dynamic scene and every static mask, we can reconstruct a series of time-varying images of the dynamic scene, given the dynamic scene is repetitive or reproducible. Experimentally, we conducted 2D and 3D imaging on a rotating chopper with a speed of 4800 revolutions per minute (rpm), and imaging speeds up to 2,000,000 fps. It is believed that this technology not only opens up a novel application direction for SPI, but also will provide a powerful solution for high-speed imaging.
科研通智能强力驱动
Strongly Powered by AbleSci AI