Due to the selective attenuation of light in water, underwater images are poorly visible and pose significant challenges in visual activities. The structural and statistical properties of different areas of degraded underwater images are damaged to different levels, resulting in an overall uneven drift of object representation and further degrading the image quality. In order to solve these problems, we introduce a method for enhancing underwater images through multi-bin histogram perspective equalization under to solve the problems caused by underwater images. We estimate the degree of feature variation in each image region by extracting the statistical features of the image and using this information to control feature enhancement to achieve adaptive feature enhancement, thereby improving the visual effect of degraded images. We first design a vibration model that exploits the difference between data elements and regular elements to improve the color correction performance of the linear transformation-based sub-interval method. In addition, a multiple threshold selection method was developed that adaptively selects a set of thresholds for interval division. Finally, a multi-bin sub-histogram equalization method is presented, which performs histogram equalization in each sub-histogram to improve image contrast. Underwater imaging experiments in various scenarios show that our method significantly outperforms many state-of-the-art methods in terms of quality and quantity.