Abstract: A synthetic aperture sonar image enhancement method based on Retina Reflectance Enhancement and variational decomposition is proposed to address the difficulty of information discrimination caused by unequal illumination in synthetic aperture sonar images. Firstly, an objective function is established through texture, structural prior, and fidelity terms, and the SAS image is decomposed to obtain estimates of the illumination and reflection components; then, the decomposed reflection components are denoised and enhanced; finally, the intensity value of the illumination component is balanced, and gamma transformation is applied to enhance the illumination component to ensure the overall enhancement effect of the image. In addition, a comparative analysis was conducted with existing classical algorithms from both subjective and objective evaluation perspectives. The experimental results show that compared with other algorithms, the algorithm proposed in this paper has a significant visual effect in processing different synthetic aperture sonar images. The processed images maintain the naturalness of the images and have obvious advantages in three aspects: contrast preservation, equalization effect, and distortion reduction.