Underwater images often suffer from challenges such as color distortion

reduced contrast

and loss of fine details due to water absorption and the scattering effect of suspended particles. The feature extraction and matching processes in underwater simultaneous localization and mapping (SLAM) systems are significantly impacted by these issues. An image enhancement algorithm for the front end of underwater visual SLAM based on multiscale fusion and detail highlighting is proposed to overcome these issues. Initially

a color correction method employing refined color channel compensation is introduced to correct the color bias in the underwater image. Subsequently

the contrast of the color-corrected underwater image is enhanced using an exposure fusion framework. The color-corrected image and the contrast-enhanced image are then fused at multiple scales. Lastly

the details of the fused image are accentuated through the application of an unsharpened mask

thereby achieving an image with superior visual effects. The experimental results demonstrate that our algorithm significantly improves the quality of processed underwater images

such as the color balance

the contrast and fine details

compared to other algorithms. Additionally

it increases the number of feature points and feature matching points

enhancing the feature extraction and matching of underwater vision SLAM.