Digital segmentation method and system

A digital, segmentation point technology, applied in character and pattern recognition, instruments, computer parts, etc., to achieve the effect of improving accuracy and good segmentation effect
CN104636744AInactive Publication Date: 2015-05-20方正国际软件(北京)有限公司 +1
4 Cites 7 Cited by

Patent Information

Authority / Receiving Office
CN · China
Current Assignee / Owner
方正国际软件(北京)有限公司
Publication Date
2015-05-20
Estimated Expiration
Not applicable · inactive patent

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention discloses a digital segmentation method and system. The method includes: subjecting an image to be segmented, to connected domain analysis, merging connected domains according to a positional relation, calculating height of each character in the image to be segmented, determining a touching connected domain to be segmented, in the merged connected domains according to the heights of the characters, building a segmentation path according to background feature points of the touching connected domain, and performing primary segmentation of the touching connected domain according to the segmentation path. If the image subjected to primary segmentation still has touching connected domains, secondary segmentation is performed according to the widths of the characters; whether or not an abnormal segmented area exists in segmentation results or not is judged; if yes, the abnormal segmented area is subjected to combination and re-segmentation. The digital segmentation method and system has the advantages that segmentation is effective and the quantity of touching characters to be segmented is not limited.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention relates to the technical field of character processing in images, in particular to a digital segmentation method and system in images. Background technique

[0002] When recognizing the text in the area to be recognized in the image, due to the influence of many factors such as the quality of the image acquisition equipment, the thickness of the text ink, and the binarization effect, the final binary image of the area to be recognized will appear sticking, fracture, etc. This will greatly affect the recognition accuracy of OCR. In order to obtain more accurate recognition results, characters need to be segmented before being sent to OCR. The connection between characters is very complicated. Firstly, the situation of character adhesion is very different. Secondly, characters may be broken. In addition, the number of character adhesion is not fixed.

[0003] In order to solve the above problems, most of the traditional methods treat adhe...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More