Edge extraction method and device

Abstract

The disclosure relates to an edge extraction method and device, and belongs to the field of image processing. The method comprises the steps that an edge image of a target object is acquired, and the edge image comprises multiple candidate edge lines; a first candidate edge line meeting the preset conditions is extracted from the multiple candidate edge lines; the first candidate edge line is extended so that an edge line grid is acquired; as for all the straight lines in the edge line grid, the characteristic values of the straight lines are calculated according to the pixel value of each line segment on the straight lines and the pixel value of the line segment arranged in the same position with each line segment on the straight lines in the edge image; and when the characteristic values of the straight lines are greater than a preset threshold value, the straight lines are confirmed to be the edge lines of the target object. Accuracy of edge extraction is enhanced, and an enclosed contour can be formed by the multiple confirmed edge lines. When three-dimensional model reconstruction is performed according to the multiple edge lines, the reconstruction effect can be enhanced.

Description

technical field [0001] The present disclosure relates to the field of image processing, in particular to an edge extraction method and device. Background technique [0002] With the development of information technology, building 3D model reconstruction has become an important means to obtain building structure information, and has a wide range of applications in the fields of urban planning, communication facilities construction and digital city construction. In order to reconstruct the 3D model of the building, it is necessary to extract the edge of the building first. [0003] When extracting the edge of a building, you can first acquire the image of the building, denoise the image to obtain a grayscale image, use Sobel (Sobel) or Canny and other operators to extract the edge of the grayscale image, and use it as the edge of the building edge. [0004] In the process of realizing the present invention, the inventors found that there are defects in related technologies, ...

AI Technical Summary

Problems solved by technology

[0004] In the process of realizing the present invention, the inventors found that there are defects in related technologies, for example: most of the edges extracted by the above method are short line segments, it is difficult to form a complete closed contour, and the short line segments include many small fluctuations, and the edges The accuracy of the extraction is poor

When 3D model reconstruction is performed from these short line segments, the reconstruction effect is very poor

Images