The invention aims to provide a contour detection method based on scale clue suppression. The contour detection method comprises the following steps: A, inputting a to-be-detected image subjected to gray level processing, presetting suppression intensity, and presetting a plurality of direction parameters uniformly distributed on the circumference, performing two-dimensional Gaussian first-order conducting function filtering to obtain classical sensory wild stimulation response of each pixel point; B, presetting a Gaussian differential function, performing the normalization processing, and obtaining a normalized Gaussian differential filtering function; C, calculating the normalized Gaussian differential filtering value of each pixel point according to the normalized Gaussian differentialfiltering function; D, presetting a high-scale value and a low-scale value of a scale clue, presetting a threshold value, and determining the scale clues of all the pixel points; E, respectively calculating the suppression response of each pixel point; subtracting the classic feeling field stimulation response of each pixel point from the suppression response of the pixel point and processing theresponse to obtain the final contour value of each pixel point. The method has the characteristics of good simulation effect and high contour recognition rate.