Underwater image sharpening method based on local sog

Abstract

The invention discloses a method for clarifying underwater images based on local SoG, which includes the following steps: assuming the underwater image is the original image, using the boundary value of the original image to perform boundary expansion to obtain a filled image; then, taking a local A window with a domain size of N×N pixels. Use this window to translate in each channel of the image after boundary expansion, estimate the light source color in each local domain, and use the light source color of the local domain as the light source color of the center pixel of the window. eλ; and through the attenuation characteristics between the light source color eλ and the original light source color, the transmittance tλ and the corresponding variance σλ of the three color channels are respectively calculated; finally, the transmittance with the largest variance σλ is subjected to a cyclic block operation to obtain background light to get the final clear image. Through a large number of experiments, it has been proved that the algorithm proposed by the present invention has very good performance in improving image clarity and restoring image color.

Description

technical field [0001] The invention relates to a computer image processing method, in particular to an underwater image clearing method. Background technique [0002] Underwater images play a very important role in the fields of marine energy exploration, marine environmental monitoring and protection, and marine military. [1] . However, due to the special underwater environment, it is difficult to obtain high-quality underwater images. Organic matter dissolved in water and particles suspended in water will absorb and scatter light, resulting in a decrease in image contrast, a reduction in the visible range of the scene, and a reduction in image quality [2] . Therefore, the problem of underwater image clarity needs to be solved urgently in the field of computer vision applications and digital image processing. [3] . [0003] Since underwater images and foggy images have similar imaging models and image features, many scholars use dark channel prior-based dehazing metho...

AI Technical Summary

Problems solved by technology

But in the underwater environment, since the attenuation of light in water is related to the wavelength and propagation distance, the estimated light source color should be attenuated, that is to say, the light source color at each point in the scene is different, which is different from the Shade of Gray algorithm estimated The globally unique light source color results in conflicting

Images