- No Data
English
- English
- 简体中文
- 繁体中文
- Français
- Русский
- Italiano
- 日本語
- 한국어
- Português
- Deutsch
- Español
- Türkçe
- Ελληνικά
- Nederlands
- Tiếng Việt
- Polski
Single pixel imaging (SPI) is a novel computational imaging technique that has been widely studied in recent years. This technology only uses single pixel detectors without spatial resolution to obtain spatial information of targets.
It has unique advantages and compensates for the shortcomings of traditional imaging technologies based on array detectors, such as relatively immature or expensive array detectors in remote sensing and non visible light imaging.
The imaging speed of SPI is always limited by the refresh rate of the spatial light modulator. For example, a typical digital micro mirror device (DMD) has a maximum refresh rate of 22kHz in binary mode. This limitation makes the implementation of real-time SPI difficult.
Researchers led by Professor Han Kai from the University of National Defense Technology of China (NUDT) are interested in single pixel imaging and fiber laser arrays. They proposed an effective SPI scheme using a phased fiber laser array and an untrained deep neural network.
Their research on efficient single pixel imaging based on compact fiber laser arrays and untrained neural networks is published in Frontiers of Optoelectronics.
Fiber lasers are arranged in a compact hexagonal structure and coherently combined to generate an illumination light field. By utilizing high-speed electro-optic modulators in each individual fiber laser module, randomly modulated fiber laser arrays can quickly project speckle onto objects of interest.
In addition, incorporating untrained deep neural networks into the image reconstruction process to improve the quality of reconstructed images.
Given its high transmission power (~kW) and potential for fast modulation (~MHz), researchers predict that the SPI scheme is expected to be applied in remote sensing and object detection.
Source: Laser Net