A short step affine transformation Sobel algorithm based image edge detection in low illumination

An object within low illumination image is difficult to distinguish and its contour needs fast computing speed in real-time visual detection systems. Aiming at above problem, a short step affine transformation (SSAT) Sobel algorithm for image edge detection is represented. First, static image needs to do affine transformation. The corresponding gradient image will be obtained after transformed image with partial derivative calculation. Then using parameter model with short step constraint, it will get the gradient different image. The last operation is to select an appropriate threshold to segment the gradient different image. Compared with one order and two order edge algorithms, SSAT Sobel algorithm can guarantee to get the right edge of image as well as meeting the real time requirements of the visual system.