The invention relates to a supercavity image enhancement method, which comprises the following steps of: 1, carrying out graying processing on a supercavity color image to obtain a gray level distribution histogram of the supercavity image, and denoising the gray level image by utilizing a Gaussian smoothing median filtering method; 2, histogram equalization processing is carried out on the de-noised gray level distribution histogram in the step one; 3, judging whether the acquisition of the navigation body in the vacuoles on the edges of the vacuoles has interference or not, if the acquisition of the navigation body in the vacuoles has interference, corroding the interference in the vacuoles by using a morphological expansion corrosion algorithm, and if the acquisition of the navigation body in the vacuoles has no interference, executing step 4; 4, performing Roots algorithm edge detection on the processed image, and extracting the edge contour of the vacuoles; and 5, carrying out data fitting by utilizing the coordinate data extracted from the edge curve, finishing mathematical modeling of the supercavity edge, and obtaining parameter data of the cavity shape. According to the method, the cavitation interference caused by complex environmental factors can be eliminated, and the supercavitation shape can be accurately obtained.