Target detection method and system based on gradient difference contrast and anisotropic filtering

The invention relates to a target detection method and system based on gradient difference contrast and anisotropic filtering. The method comprises the following steps: S1, setting parameter information; s2, carrying out defogging processing on the original image to obtain a defogged image; s3, calculating gradient differences in all directions by constructing an anisotropic diffusion model, and obtaining a sharpened image by using a Gaussian filtering model; s4, performing differentiation processing on the sharpened image from the orthogonal region and the diagonal region, and performing fusion and superposition of gray difference contrast of the orthogonal region and the diagonal region to obtain a saliency image; s5, calculating a grayscale adaptive segmentation threshold value of the saliency image, and obtaining an adaptive segmentation threshold value after segmentation; and S6, performing connected region processing on the segmented adaptive segmentation threshold to obtain target position information.