Improving hyperspectral imaging using a lensless camera

Hyperspectral imaging, or HSI, is an imaging technique that records images at distinct wavelengths and continuously samples across a broad spectrum, offering a comprehensive analysis of objects in terms of their spectral properties. While enabling more accurate detection, classification, and segmentation of target objects than regular color photography, HSI systems typically require complex, bulky, and costly optical systems.

Conventional HSI systems often involve sequential scanning in the spatial or spectral domain, creating a complex trade-off between spatial, spectral, and temporal resolution. Kim et al. developed a lensless snapshot hyperspectral camera based on computational imaging that is compact and uses a low-cost hardware configuration.

“Our camera pairs a phase mask with a linear variable filter that yields a continuous transmission spectrum allowing us to adjust the spatial/spectral resolution according to our needs by dividing the point-spread function pattern,” said Seung Ah Lee. “Our camera also offers simplicity, snapshot imaging capabilities, and the advantage of customization of components without the need for significant modifications.”

Hyperspectral images ranging from 410 to 800 nanometers were obtained with the prototype device.

“We anticipate our camera will facilitate easy spectral data acquisition across a wide range of applications,” said Lee.

Medical imaging, such as used in disease diagnosis and image-guided surgery, may benefit significantly from lensless HSI. By adjusting filter bandwidth or combining filter sets, additional contrast for fluorescence imaging can be achieved. A compact hyperspectral endoscope could be built to distinguish between different tissue types. Other applications of the technology include remote sensing, food quality estimations, and astronomy.

Source: “Aperture-encoded snapshot hyperspectral imaging with a lensless camera,” by Taeyoung Kim, Kyung Chul Lee, Nakkyu Baek, Hyesuk Chae, and Seung Ah Lee, APL Photonics (2023). The article can be accessed at https://doi.org/10.1063/5.0150797 .