Point spread function measurements for underwater imaging: an analysis of wavelength-specific behavior for image...

Area of Science:

• Optical oceanography
• Image processing
• Remote sensing

Background:

• Underwater optical imaging, particularly in coastal areas, faces challenges with reduced image quality due to light scattering.
• The increasing use of hyperspectral and multispectral imaging necessitates a deeper understanding of the point spread function (PSF) across different wavelengths.

Purpose of the Study:

• To experimentally measure and analyze the wavelength-specific point spread function (PSF) in various turbidity conditions.
• To evaluate the impact of wavelength-dependent PSFs on underwater image restoration.
• To assess the applicability of model-based PSFs for effective image restoration across different wavelengths.

Main Methods:

• Conducted extensive laboratory measurements of the PSF at multiple wavelengths (450, 500, 550, 600, 650 nm) in controlled turbidity.
• Utilized deionized water with Maalox as a scattering agent and natural water samples.
• Employed a high dynamic range (HDR) camera calibration for accurate PSF measurements across a wide range of magnitudes (>6 orders).

Main Results:

• Observed significant differences in PSF shapes across different color channels, particularly with increasing turbidity (Maalox concentration).
• Image deconvolution demonstrated the critical importance of using wavelength-specific PSFs, especially for correcting longer wavelength channels.
• Validated that properly scaled, model-based PSFs derived from single-wavelength measurements can effectively restore images at other wavelengths over long scattering distances.

Conclusions:

• Wavelength-specific point spread functions are essential for accurate underwater image restoration, especially in turbid coastal waters.
• The findings support the use of scaled, model-derived PSFs for broad-spectrum image correction, simplifying practical applications.
• This research provides a foundation for improving the quality and reliability of underwater optical imaging systems.