A method for layer-by-layer segmentation of distorted inkjet characters
A coding and character technology, applied in the field of layer-by-layer segmentation of distorted inkjet characters, can solve problems such as no effective solution, and achieve the effect of high inkjet character segmentation accuracy and robustness
Active Publication Date: 2022-04-22
XI AN JIAOTONG UNIV +1
View PDF13 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Currently, there is no effective solution to this problem
Method used
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
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment
[0060] The present invention relates to text segmentation algorithms such as figure 1 As shown, the original input image as figure 2 , the position and angle of the workpiece are randomly placed in the field of view. Firstly, the input image is centered and corrected, and a threshold value of 25 is selected to binarize the input image to separate the workpiece to be measured from the background. Then, the coordinates of the center of gravity of the binarized image are calculated, and the translation correction amount is obtained by comparing the center coordinates of the image (such as image 3 ); then extract the grayscale interval of the relevant pixels of the inkjet text [43,81], and obtain the interval threshold image (such as Figure 4 ), and filter the connected domains that are too large and too small to obtain a text binary image with a small amount of noise. By calculating the grayscale projection of the text at each angle, find the angle corresponding to the projec...
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 PUM
Login to View More Abstract
The invention discloses a method for layer-by-layer segmentation of distorted code-spraying characters. The method combines three ideas of character segmentation: connected domain box selection method, gray-scale projection valley detection method, and average segmentation method; character connected domains are divided from coarse to fine Three subdivisions, in which the "connected domain frame selection method" solves the problem that the "gray projection valley detection method" is easily interfered by the gaps in the dot matrix, and the "gray projection valley detection method" solves the problem that the "average segmentation method" has long segmentation distortion. The problem of judging the cumulative number of characters in the character area. Compared with the single character segmentation method, the present invention shows stronger robustness in dealing with a certain degree of distorted sprayed code character segmentation problem, and at the same time, the subdivision structure of sub-regions limits the scope of action of the three segmentation methods, and will not Because the method itself brings additional segmentation errors, it brings a higher accuracy rate of inkjet character segmentation.
Description
technical field [0001] The invention belongs to the field of image processing and optical character recognition, and in particular relates to a layer-by-layer segmentation method for distorted code-spraying characters. Background technique [0002] The coded characters of a certain workpiece are important identity information corresponding to specific workpieces such as product parameters and quality indicators. At present, the recognition of inkjet characters mainly relies on manual inspection. The difficulty in the application of the general character recognition method to the coded characters is mainly character segmentation. Since the coded characters are composed of discrete dot matrixes, the character interval characteristics are not obvious, and the dot matrix gaps in the font will cause the gray projection value in the corresponding direction to decrease. Cause interference; in addition, when the coding direction cannot be perpendicular to the surface of the workpie...
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 Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06V10/26G06V30/24
CPCG06V30/153G06V10/267
Inventor 黄军辉王昭杨程祁苗伟高建民张厅方
Owner XI AN JIAOTONG UNIV