A Method for Identifying Sources of Document Images Based on Continuity of Straight Lines

A document image, discrimination method technology, applied in character and pattern recognition, instruments, computer parts and other directions, can solve the problems of misleading results, inaccurate classification, etc. Effect

Active Publication Date: 2017-02-15
XI AN JIAOTONG UNIV
View PDF1 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, filtering is difficult to obtain an ideal result when processing binary images with less information, and some raster images have halftones to represent gray patterns, which further misleads the filtered results and eventually leads to incorrect classification. precise

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A Method for Identifying Sources of Document Images Based on Continuity of Straight Lines
  • A Method for Identifying Sources of Document Images Based on Continuity of Straight Lines
  • A Method for Identifying Sources of Document Images Based on Continuity of Straight Lines

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0028] The present invention will be further described in detail below in conjunction with the drawings and specific embodiments.

[0029] Such as figure 1 As shown, a method for identifying the source of document images based on straight line continuity of the present invention mainly includes three parts: a straight line segment and isolated noise point detection part, a calculation base straight line length part, and a straight line continuity feature structure and classification part. The specific method is as follows:

[0030] Step 1: Perform an edge extraction operation on the input binary image, and output the edge image with the filling part removed;

[0031] Step 2: Detection of straight line segments and isolated noise points: For the edge image output in step 1, first use a 3*N straight line detection template to detect a straight line segment with a pixel length of N. The specific detection process is: use one and 3*N The window of the same size of the straight line dete...

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

No PUM Login to view more

Abstract

A document image source discrimination method based on the continuity of straight lines. First, the edge extraction operation is performed on the input binary image, and then the straight line segment detection in the horizontal and vertical directions is performed using a 3*N straight line detection template, and obtained by searching The complete straight line segment and the isolated noise points in the local area at both ends of the straight line segment; then a two-way extended search is performed on the detected straight line segment to obtain the length of the base straight line corresponding to the straight line segment; finally, classify according to the length of the base straight line, and calculate The ratio of the length of the straight line segment in each class to the length of the base line is used as a feature, and the ratio of the number of isolated noise points to the number of straight line segments is added as an additional feature and then input into the trained SVM classifier for classification, and finally the category of the output image; The invention aims at the deficiencies and gaps in the binary document image source discrimination method, and can quickly distinguish most document images containing straight lines on the basis of ensuring no misjudgment.

Description

Technical field [0001] The invention relates to the technical field of a method for discriminating the source of a document image, in particular to a method for discriminating the source of a document image based on the continuity of a straight line. Background technique [0002] Document images can be divided into scanned images and raster images according to their origin. The image storage can be divided into color images, grayscale images and binary images according to the amount of information stored in each pixel. [0003] At present, the research on algorithms for identifying the source of document images can be divided into three categories: [0004] One is a method based on document tilt detection, which is mainly aimed at the situation where the image of the scanned document in the early period often appears to be tilted. This type of method generally first obtains a whole line of text or the blank area between text lines, and calculates the angle between the text line and...

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
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/00G06K9/62
Inventor 宋永红郁冲张元林
Owner XI AN JIAOTONG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap