Short-wave infrared (SWIR) hyperspectral imaging is a cutting-edge technology that captures three-dimensional (3D) spectral-spatial information in the SWIR range. This capability allows for the identification and characterization of materials and targets based on their spectral features, making it useful in a variety of fields such as chemical analysis, material identification, agriculture, food industry, quality control, and military reconnaissance. However, the high cost of indium gallium arsenide (InGaAs) focal plane arrays (FPAs) has limited the widespread adoption of SWIR hyperspectral imaging. Recently, a research team led by Professors Sun Baoqing and Gao Yuan from the School of Information Science and Engineering and the Ministry of Education Key Laboratory of Laser and Infrared System Integration published a study titled "Quantum dot-enabled infrared hyperspectral imaging with single-pixel detection" in the prestigious journal (Light: Science & Applications (IF: 19.4)). This study provides a potential solution to the cost issue (article link: https://www.nature.com/articles/s41377-024-01476-4).

Advancements in algorithms and computational power have increased interest in computational spectral reconstruction using broadband light encoding. Colloidal quantum dots (CQDs) for spectral encoding and reconstruction were first proposed by Professor Bawendi, a Nobel Prize in Chemistry laureate in 2023, in 2015. CQDs can continuously modulate their absorption characteristics by adjusting their size and chemical composition, covering wavelengths from ultraviolet to mid-infrared. This flexibility makes CQDs ideal for achieving broadband spectral modulation and encoding. Additionally, CQDs have distinct excitonic absorption features, offering higher spectral encoding randomness and efficiency compared to traditional color filters.

In this study, the research team synthesized a series of monodisperse lead sulfide (PbS) quantum dots covering the SWIR range by controlling the synthesis conditions. By managing the surface characteristics of CQDs and the evaporation rate of the solution, they created SWIR filters based on CQD self-assembly structures, enhancing the CQD's infrared absorption efficiency. This resulted in a set of quantum dot filters with varying transmission characteristics, enabling spectral encoding across the SWIR range.

Figure 1: Physical and transmission electron microscope images of colloidal quantum dot filters

Hyperspectral images are processed and analyzed as a 3D (x, y, λ) data cube, where x and y represent the two spatial dimensions of the scene, and λ represents the spectral dimension. Traditional hyperspectral imaging methods usually acquire this data cube through spatial or spectral scanning. To avoid the high cost of 2D SWIR sensors and complex wavelength selection components, this study uses CQDs and a digital micromirror device (DMD) to encode spectral and spatial information (Figure 2). By leveraging a single-pixel detector and compressed sensing algorithms, the transmission spectra of CQD filters are correlated with the projection patterns generated by the DMD, yielding high-resolution SWIR hyperspectral images. Each pixel contains complete spectral characteristics, enabling coordinated reconstruction of spectral and spatial dimensions using single-pixel detection principles.

Figure 2: Schematic of the quantum dot SWIR hyperspectral imaging system

The research team employed self-assembled colloidal quantum dot filters and digital micromirror devices to encode SWIR spectral and spatial information. They utilized single-pixel detection principles for coordinated reconstruction of spectral and image data. This approach has produced high-quality SWIR hyperspectral imaging results, with spectral and spatial data accurately matching those obtained by reference instruments. The successful implementation of quantum dot-based infrared hyperspectral single-pixel imaging highlights the potential for low-cost, miniaturized spectral encoding chips in hyperspectral imaging systems. The combination of quantum dots and single-pixel detectors effectively reduces system complexity and cost, potentially promoting broader civilian applications of SWIR hyperspectral imaging technology.

Professors Sun Baoqing and Gao Yuan from the School of Information Science and Engineering of Shandong University and the Key Laboratory of Laser and Infrared System of the Ministry of Education are the corresponding authors. Meng Heyan, a Ph.D. student from the class of 2023, is the first author. This research was supported by the National Key Research and Development Program of China, the National Natural Science Foundation of China, the Shandong Taishan Scholar Young Expert Program, the Young Interdisciplinary Science Innovation Group of Shandong University and the Qilu Young Scholar Program of Shandong University.