Underwater Image Enhancement Methods and Applications

Underwater imaging presents unique challenges arising from the interaction of light with the aquatic medium. Absorption and scattering of incident light reduce contrast, distort colour and degrade fine details, particularly in turbid or deep-water environments. A range of methods has been developed to address these impairments, including physics-based restoration that inverts image formation models, non-physical approaches that apply global and local contrast enhancement, and hybrid schemes that integrate optical models with data-driven techniques. Polarimetric recovery exploits the state of polarisation to discriminate between direct and backscattered light, histogram-based methods adjust channel distributions to correct colour cast and visibility, and Retinex-inspired algorithms emulate human vision to enhance local contrast. More recently, deep learning architectures—particularly residual networks—have leveraged synthetic data and adversarial training to jointly correct distortion, sharpen detail and restore natural colour balance. Together, these methods have found diverse applications in marine biology, archaeology, offshore engineering and robotics, enabling improved object detection, habitat mapping and visual navigation in subaqueous environments.

Research from Nature Portfolio

One foundational study introduced a polarimetric recovery method that combines histogram stretching of orthogonally polarised images with traditional polarimetric reconstruction. By preserving polarisation relations while equalising intensity distributions, this approach yields improved visibility and colour fidelity in dense turbid media. Experimental results demonstrate its superiority over conventional polarimetric methods, with particular effectiveness in scenarios of strong backscatter and signal attenuation.

Underwater Image Enhancement Methods and Applications publication trend

The graph below shows the total number of articles in underwater image enhancement methods and applications across all publications each year (not limited to Nature Index journals).

Technical terms

Scattering: Deflection of light by suspended particles in water that reduces image clarity.

Absorption: Attenuation of light intensity as it travels through water, leading to colour shifts and loss of brightness.

Backscatter: Portion of scattered light directed toward the camera, causing veiling glare and contrast loss.

Polarimetric imaging: Technique that captures the polarisation state of light to separate direct scene radiance from scattered components.

Retinex framework: Computational model inspired by human vision that enhances local contrast by comparing pixel intensities across scales.

Histogram stretching: Global intensity transformation that redistributes pixel values to span the available dynamic range.

Deep residual network: Neural network architecture using skip connections to ease training of very deep models for image restoration.

References

1. Polarimetric image recovery method combining histogram stretching for underwater imaging. Scientific Reports (2018).
2. Underwater image enhancement via extended multi-scale Retinex. Neurocomputing (2017).
3. Underwater image and video dehazing with pure haze region segmentation. Computer Vision and Image Understanding (2018).
4. Underwater Image Enhancement With a Deep Residual Framework. IEEE Access (2019).