Complementary metal-oxide-semiconductor (CMOS) sensors are found in many imaging systems, such as cameras, and can be tuned to receive various ranges of wavelengths of light. In low-visible-light conditions, a camera with a CMOS sensor tuned to visible light, as is common in many portable electronic devices, may struggle to produce clear or detailed images. Likewise, even though a camera with an infrared-tuned CMOS sensor may detect additional information unavailable to the visible-light CMOS sensor in low-visible-light conditions, the resulting infrared-only image would lack an accurate color representation of the scene. Nevertheless, an imaging system featuring a machine-learning computer-aided vision system configured to blend color data from a visible-light-tuned CMOS sensor and morphological or topographical data from an infrared-tuned CMOS sensor can produce clear, colorful, and detailed images.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Murugan, Kavinaath and Kandibanda, Aaditya, "Improving Low-Light Imaging by Using Infrared and Software Interpolation", Technical Disclosure Commons, (May 02, 2019)