Multiple linewidth self-adapting preliminary vectorization method in vectorization process of engineering drawing

Abstract

The invention relates to chain including the location of the central axis point coordinates and width information; D. adop The invention publishes a more linewidth adaptive initial vector method in the course of works drawings vector, steps are: A. tracks line composed of black pixel regional of raster map by the method of determining the key points on the client, for the initial point; B. uses the method of linewidth automatical estimation and method of automatically selecting the largest tracking step based on linewidth on the client ,for the initial linewidth and the largest tracking step; C.adopts method of adaptive tracking based on the size of linewidth on the client ,for the central axis point ts automatic judgment on merits of the axis point chain based on length of the axial chain at the client, and judges whether it finished scanning grating map, put the output data to the server after scanning. Using the above method invented to ameliorate the results of the initial vector, improve the accuracy of the initial vector, expands the scope of vector grating drawings and accelerates to a certain extent, the computational speed of the initial vector.

Description

Technical field: [0001] The invention relates to a vectorized recognition method for engineering drawings, in particular to a vectorized method for complex raster images with different line widths. The method can be widely used in the fields of engineering drawing recognition, image graphic symbol detection and recognition, and industrial automation detection and other fields. Background technique: [0002] In today's information society, although CAD technology has been widely used in the field of industrial design, there are still a large amount of drawing information stored, managed and applied in the form of paper and pictures. These engineering drawings are often accumulated by a system or industry for a long time, completely preserving the valuable experience in the past, and have a good reference for future design. A considerable part of them will be modified and reused in future design and manufacture. In order to reuse these drawings, it is an inevitable developme...

AI Technical Summary

Problems solved by technology

At present, the research on vectorization is basically carried out under the framework of these methods, but for the noise interference of the image, only the method of discrete image tracking can achieve better results, but the method of discrete image tracking is effective for raster images. The width requirement is stricter, which makes some drawings with thinner line widths have poor initial vectorization results

At the same time, due to the method of fixing the maximum tracking step size, the accuracy of the preliminary vectorization results of engineering drawings with large line width changes is low

Images