An illumination balance algorithm based on improved affine shadow formation model for underwater image

The uneven illumination distribution in underwater visual inspection will lead to the difficulty of extracting texture features. The underwater image illumination balance while keeping the texture details has been one of the key issues in underwater visual inspection. Aimed at this problem, a novel illumination balance algorithm based on improved affine shadow formation model is proposed in this study. In the proposed approach, the linear spatial filter is used to obtain the light intensity distribution of an image, and the original image is divided into a series of small strips of pixels based on the light intensity distribution. Then the illumination balance of the image is carried out based on an improved affine shadow formation model. The experimental results show that the proposed approach can deal with the uneven illumination problem in underwater image, and keep the texture details effectively, which is very important for the subsequent processing and analysis for underwater images.