Joint edge detector based on Laplacian pyramid

In order to capture the details of the edge, continuous, thin and noise free, this paper describes an improved edge detection based on Laplacian pyramid. We studied the edge detection based on Laplacian Pyramid (LP edge detector) in detail, proposed Joint edge detector which is consists of the Laplacian sharpening and Laplacian pyramid. This method is equivalent to that the Laplacian sharpening was followed by a Gaussian impulse response low-pass filter and multiscale image decomposition. Laplacian sharpening enhanced image details, and the two follow up steps improved the stability of the noise. The experimental results showed that the method inherits the advantages of multiscale edge detection, plays to the advantages of Laplacian sharpening. The method improves the ability to capture the details of the edge, and eliminates the noise.