Pixel-level tunable video hyperspectral intelligent imaging technology can be used in many fields

The Light Conference 2019 international conference was successfully concluded at the Changchun Institute of Optics, Fine Mechanics and Physics. Among the invited expert reports of this conference, the "Pixel-level Tunable Video Hyperspectral Intelligent Imaging Technology" made by Associate Researcher Hu Changhong of the Changchun Institute of Optics, Fine Mechanics and Physics attracted widespread attention. In April 2019, the project team of Han Chengshan and the project team of Wang Xiaoyi of the Changchun Institute of Optics, Fine Mechanics and Physics jointly developed a prototype of a video hyperspectral camera based on pixel-level spectroscopy technology with completely independent intellectual property rights. This achievement can be used in civil disaster reduction, land resources, precision agriculture/forestry, food and drug testing, industrial testing (product defects), and cultural relics identification.

Video Hyperspectral Camera Principle Prototype

The project is based on the Jilin Province Science and Technology Research Project "Research on Video Hyperspectral Imaging Technology". According to the project leader Hu Changhong, the camera is a staring hyperspectral camera with the characteristics of light weight, small size, simple structure, high stability, and no need for data calculation. It can achieve high frame rate staring imaging and hyperspectral imaging that can obtain multiple spectral bands in a single exposure, and has hyperspectral video imaging capabilities with high spatial resolution, high temporal resolution and high signal-to-noise ratio.

At present, my country's hyperspectral imaging technology is usually based on prism/grating spectrometry, interference or linear gradient filter and other spectroscopic systems, and works in a push-sweep or swing-sweep mode, which cannot achieve hyperspectral video-level imaging. Video hyperspectral imaging technology based on pixel-level spectrometry technology is mastered by a few companies such as IMEC in Europe, Silios in France, and Pixelteq in the United States, while there are few reports of similar products in China. After hard work, the Han Chengshan project team and the Wang Xiaoyi project team of the Changchun Institute of Optics, Fine Mechanics and Physics have broken through the technical difficulties of hyperspectral pixel-level spectrometry, pixel-level optical-mechanical assembly and integration, high-frequency and high-speed electronics design, and spectral data inversion and restoration, and successfully developed a video hyperspectral imager based on pixel-level spectrometry technology. Compared with similar foreign products, the performance indicators are comparable, successfully breaking the monopoly of similar foreign technologies and enriching my country's imaging methods in the field of hyperspectral remote sensing.

Currently, the project team has used this achievement to conduct imaging recognition experiments on camouflage and plants, with good results. The next step is to transform the product into achievements.