Active and low-cost hyperspectral imaging for spectral analysis in low lighting environment

Hyperspectral imaging can capture information beyond conventional RGB cameras; thus, it has many applications, such as material identification and spectral analysis. However, like many camera systems, most of the existing hyperspectral cameras are still passive imaging systems: they require external light sources to illuminate the objects to capture the spectral intensity. As a result, the collected images highly depend on the environment lighting, and the imaging system cannot function in a dark or low-lighting environment. This work develops a prototype system for active hyperspectral imaging, which actively emits different single-wavelength lights at different frequencies when imaging. This concept has several advantages: first, using the controlled lighting, the magnitude of the individual bands is normalized to extract reflectance information; second, the system is capable of collecting information at the desired spectral range by tailoring the light sources; third, an active system is mechanically easier to make, since it does not require complex band filters as used in passive systems; last, such a system may work under low light or dark environments, which greatly facilitate underground/subsurface sensing applications such as borehole based mining exploration. This prototype is achieved by using an array of low-cost and single-wavelength LED (Light Emitting Diode) lights, a remote control module controlling the LED illuminator, and the shutter of a full spectrum camera. We demonstrate that such design is feasible and could yield informative hyperspectral images for spectral analysis and machine learning-based object identification in low light or dark environments, having great potential to benefit both the academic and industry such as in geochemistry, earth science, subsurface energy, and mining.