Underwater hyperspectral imaging bioinspired by chromatic blur vision

Area of Science:

• Optics
• Biomimetics
• Image Processing

Background:

• Conventional underwater hyperspectral imaging (UHI) systems face limitations due to bulky platforms and complex optics, hindering efficiency.
• Existing methods struggle with stable imaging and rapid data acquisition in dynamic underwater environments.

Purpose of the Study:

• To develop a novel, efficient underwater hyperspectral imaging system inspired by cephalopod vision.
• To leverage chromatic blur for rapid spectral image reconstruction with minimal data input.

Main Methods:

• Designed a bionic lens that enhances chromatic blur for band discrimination.
• Collected underwater chromatic blur datasets for network training.
• Reconstructed 30-band spectral images (430-720 nm) from only three input frames.

Main Results:

• The novel UHI system successfully reconstructed hyperspectral images with high spectral discrimination.
• Achieved a 90% reduction in data sampling rate and a 99.98% reduction in exposure time (2.1 ms) compared to traditional methods.
• Demonstrated significant potential for rapid response in underwater dynamic scenario recognition.

Conclusions:

• The proposed chromatic blur vision approach offers a highly efficient and rapid solution for UHI.
• This biomimetic system expands the practical applications of hyperspectral imaging in challenging underwater environments.