Unlock instant, AI-driven research and patent intelligence for your innovation.

Character image line selecting method and device and character image identifying method and device

An image and text technology, applied in the field of text and image recognition, can solve problems such as segmentation errors, poor recognition results, wrong positions and sizes, etc.

Inactive Publication Date: 2003-04-16
CANON KK
View PDF0 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] It can be seen that due to the error of line segmentation, the original 21 valid text lines are only divided into 8 lines, and because the position and size of these line images are wrong, the recognition result is very poor

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
  • Character image line selecting method and device and character image identifying method and device
  • Character image line selecting method and device and character image identifying method and device
  • Character image line selecting method and device and character image identifying method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0021] Embodiments of the present invention will be described below in conjunction with the accompanying drawings.

[0022] Through the analysis of the original algorithm, it can be seen that when the noise in the image segment is relatively concentrated in a certain area, it will "cover" the blank pixel rows in this area. If there is a lot of noise, the gap between the "peak" and "trough" in the pixel distribution statistical diagram will be further narrowed, making it difficult to judge the position of the text line. For this reason, inventor has proposed new character image branching method ( Figure 6A ).

[0023] Such as Figure 6C As shown, the text image is input into the text image sorting device 602 by the text image input device 601 (such as a scanner, etc.) to perform the text image sorting. The character segmentation device 603 performs character segmentation on the line image. The character feature extraction and recognition device 604 performs feature extract...

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

The method includes following procedures. The segmentation procedure divides the pixels of the character image into multiple sections of the character image with certain width side by side. The procedure of the pixel distribution statistics obtain the pixel distribution in each image section. Namely, it is the number of the black pixels in each pixel row of the image section. The pixel distribution of the entire image of obtained. Namely, it is the number of the black pixels in each pixel row of the entire image. The section block forming procedure: base on the statistics of the pixel distribution in the image section and in entire image obtained in the procedure of the pixel distribution and statistics, the image section is divided into section block.

Description

technical field [0001] The invention relates to a method and device for character image line division, in particular to the line division in character image recognition. Background technique [0002] Figure 1A Represents the flow of existing text and image recognition algorithms. Figure 1B It is a structural diagram of an example of a text and image recognition device in the prior art. First at s101 character image line segmentation device 112 is divided into lines by the text image of input device 111 input (as by scanning); at s102 character segmentation device 113 is segmented with the character in each line; Character feature extraction and identification device 114 Extract the features of the segmented characters at s103, perform character matching and recognition at s104; output the recognition result at s105 by the output device 115. In the method of character image recognition, the accuracy of the image branch directly affects the accuracy of the final character ...

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 Applications(China)
IPC IPC(8): G06K9/20
CPCG06K9/00463G06V30/414
Inventor 罗兆海李毅
Owner CANON KK