Remote sensing image edge-detection based on improved Canny operator

Abstract

The edge details of the remote sensing image is key for the target detection with complex background. Traditional Canny detection operator requires human intervention, does not have the adaptive ability in the variance of Gaussian filtering and the threshold. Compound morphological smoothing replaces Gaussian filtering, can not only maintain the edge information, but also reduce noise impact. Then Otsu method can specify the threshold adaptively, the edge detected is more continuous, and can decrease the false edges. In order to extract the edges of the remote sensing image proposed an approach based on Canny edge detection operator. Finally, we refine the edge by using morphological structure element. The experimental results show that the improved Canny algorithm has a good anti-noise function and precision on image processing of remote sensing.