High-Resolution Multispectral Photovoltaic Imagers from Visible to Short-Wave Infrared

Area of Science:

• Optoelectronics
• Materials Science
• Image Processing

Background:

• Multispectral imaging is crucial for agriculture, security, and medical diagnostics.
• Traditional systems suffer from complex alignment and resolution loss due to separate sensors.
• A need exists for integrated, high-resolution multispectral imaging solutions.

Purpose of the Study:

• To develop a hardware-algorithm co-designed architecture for multispectral super-resolution imaging.
• To demonstrate a monolithic quad-spectral photovoltaic imaging platform.
• To overcome resolution degradation in planar spectral sensing.

Main Methods:

• Developed a monolithic quad-spectral photovoltaic imaging platform (640 × 512 pixels).
• Integrated visible to short-wave infrared (350–2350 nm) spectral bands using an all-polymer bulk heterojunction and colloidal quantum dots.
• Employed a super-resolution reconstruction method to restore image resolution.

Main Results:

• Achieved broadband spectral integration in a single CMOS-compatible architecture.
• Demonstrated a photovoltaic imaging platform with <1% dead pixels per channel.
• Successfully restored images to a resolution of 640 × 512 using super-resolution reconstruction.

Conclusions:

• The co-designed architecture enables simultaneous multispectral data capture and processing.
• This technology paves the way for CMOS integration and advanced multispectral imagers.
• Applications span precision agriculture, medical diagnostics, and beyond.