Abstract: In order to extract edges of target area more completely and accurately from remote sensing
images, a method of target edge extraction is proposed based on improved mathematical morphology
and modulus maxima of non-subsampled Shearlet transform. Firstly, the image is decomposed into
high-frequency components with more edges and details and low-frequency component with fewer
edges and minutiae through non-subsampled Shearlet transform. Then considering the property of
coefficients of edge points under different decomposing conditions, the modulus maximum detection is
performed for each sub-band of high-frequency components and the double-layer mask is adopted
afterwards so as to get the high-frequency edge extraction result. Moreover, the low-frequency
component is processed through the improved mathematical morphology method to get the
low-frequency edge extraction result. Finally, the above two parts are fused and the final target edge
image is obtained after removing the isolated points according to the regional connectivity. A large
number of experimental results show that, compared with Canny method and four similar edge
extraction methods, the detected edges by the proposed method are accurate, clear, complete and with
abundant details. The method has strong anti-noise performance, which lays a better foundation for the
following target feature extraction and recognition of remote sensing images.

Key words: target edge extraction, remote sensing images, non-subsampled Shearlet transform;
mathematical morphology, regional connectivity