Method and device for overlapping region segmentation based on pit point matching

Abstract

The invention discloses an overlapping region segmentation method and a device based on concave point matching. The method comprises the following steps: receiving an original grain image; marking anoverlapping target region by screening the centroid number of the same connected region in the original grain image after preprocessing; determining a segmentation target; and segmenting the overlapping target region; using at least two concave point search methods for concave point selection of the segmentation target to form an initial concave point set; setting a distance threshold of adjacentconcave points, and removing redundant concave points in an initial concave point set according to the threshold; the equal number of concave points being divided and the maximum curvature point in the concave point group being selected as the representative concave point of the concave point group. A representative concave point of a concave point group is taken as a concave point to be matched,and a concave point which is closest to the curvature of the concave point to be matched and is connected with a line in a target region and intersects with a connecting line of a corresponding centerof mass is taken as an alternative matched concave point, and the overlapping area of the received matching concave point and the nearest alternative matched concave point is divided, and all the concave points are traversed to complete the division.

Description

technical field [0001] The disclosure belongs to the technical field of crystal grain identification, and relates to a method and device for dividing overlapping regions based on pit point matching. Background technique [0002] The crystallization industry is an important part of the world economy, and crystallization products are widely used in pharmaceutical, food, chemical and other industries. The shape and size distribution of grains during the crystallization process is of great significance for the control of the crystallization process, which can avoid unnecessary product forms in time. [0003] At present, the relatively advanced measurement method in the field of crystallization process control is the monitoring method based on image processing technology. By acquiring the grain image and performing image processing to identify the grain shape and obtain the particle size distribution, the method in the image processing process is very important for the characteri...

AI Technical Summary

Problems solved by technology

[0006] In view of the deficiencies in the prior art, solve the pit matching segmentation method for overlapping grain images in the prior art, there are many false pits and the matching rules are not good, resulting in the inability to achieve accurate segmentation of overlapping grain images problem, this disclosure provides a method and device for overlapping region segmentation based on pit matching, which realizes accurate segmentation of overlapping regions of grain images

Images