Underwater Image Enhancement Method Based on Illumination Correction and Color Correction

Abstract

Due to the attenuation and scattering of light propagating in water, underwater images are subject to blurring, uneven illumination, and color distortion. To cope with these problems, an underwater image enhancement method building upon illumination correction and color correction is proposed. This method makes an effective integration of illumination correction and color correction. To commence with, the underwater fish image is deblurred by dark channel prior (DCP). Subsequent to this, contingent upon the varying brightness of the images arising from the uneven illumination, the camera response model is used to perform low illumination enhancement on the low-brightness areas, and the high-brightness areas undergo exposure reduction. As such, the generated low-brightness images are fused with the de-exposed images by the principal component analysis (PCA) fusion algorithm. Finally, based on the gray world hypothesis theory, the deblurred and illumination-corrected images proceed with the color correction to generate the final underwater enhanced image. Experiments on manifold underwater images demonstrate that the proposed method exhibits good performance in objective evaluation indexes. When juxtaposed with other methods, this method can substantiate the brightness of low-illumination areas while maintaining the brightness of well-exposed areas, and effectively solve the problem of color distortion.