<p>Due to the complexity of the underwater environment and the limitations of the equipment, it is often difficult to obtain satisfactory underwater images, and the obtained images need to be processed. Therefore, underwater image enhancement has received extensive attention. Some commonly used terrestrial image processing methods are often difficult to apply to complex underwater environments, which can cause problems of image color distortion and artifacts. With the growing interest in deep sea exploration, deep sea image enhancement has become a new research hotspot, facing unique challenges. In this paper, we experimentally analyze the differences between shallow and deep sea imaging environments and design an image enhancement method. Unlike the currently existing methods, our proposed enhancement method is applicable to both deep-sea and shallow-sea environments at the same time. The method is improved on the basis of traditional white balance by first classifying the images and then designing color compensation strategies for color compensation respectively. In addition, our adaptive contrast enhancement method adopts the idea of block division and utilizes a bilinear interpolation algorithm to process the image. It works well without introducing artifacts, even in the face of extreme darkness and interference from artificial light sources. Our method is a single-image approach and requires no specialized hardware or knowledge of underwater conditions or scene structure, making it highly practical. Finally, in order to evaluate the effectiveness of the method more comprehensively and objectively, a total of 3290 images are used for testing. The subjective, objective evaluations and detail comparisons prove that our method has obvious advantages over other methods.</p>