An image normalization method and device based on table characteristics

A normalization and image technology, applied in 3D image processing, image data processing, graphics and image conversion, etc., can solve problems affecting the bill image recognition rate, cutting out the recognition area, and inaccurate frames, etc., to improve the bill Recognition rate, improve image quality, excellent effect

Active Publication Date: 2019-06-28
厦门商集网络科技有限责任公司
View PDF4 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Using a fixed template will cause problems such as inaccurate frames and incomplete frames, and then the correct recognition area cannot be cropped, which greatly affects the bill image recognition rate

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
  • An image normalization method and device based on table characteristics
  • An image normalization method and device based on table characteristics
  • An image normalization method and device based on table characteristics

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0040] see figure 2 , a method for normalizing images based on table features, the table is composed of a plurality of non-overlapping cells, comprising the steps of: obtaining images to be recognized (such as Figure 4 shown), generate a template map corresponding to the image to be recognized (such as image 3shown), the template image is a standard template image corresponding to the image to be identified; select at least one corresponding cell from the image to be identified and the template image respectively, and select from the selected cells At least four vertices located on the outer frame are confirmed, and the vertices confirmed in the image to be identified correspond to the positions of vertices confirmed in the template image, forming at least four sets of corresponding vertex coordinates; according to the corresponding vertices The coordinates calculate the perspective transformation matrix; create a blank picture, the size of the blank picture is the same as...

Embodiment 2

[0051] This embodiment provides a device for normalizing images based on table features, including a memory and a processor, the memory stores instructions, and the instructions are suitable for being loaded by the processor and performing the following steps:

[0052] Match the cell in the image to be identified with the cell in the template image, the image to be identified is an image that needs to be automatically identified, the template image is a standard template image corresponding to the image to be identified, if the matching is successful , perform the following steps: respectively select at least one corresponding cell from the image to be identified and the template image, and confirm at least four vertices on the outer border from the selected cells, the image to be identified The vertex confirmed in the corresponding to the vertex position confirmed in the template diagram, forming at least four groups of corresponding vertex coordinates; use the corresponding v...

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 invention relates to an image normalization method based on table characteristics. The method comprises the following steps of: preparing a substrate;, determining a template graph corresponding to the to-be-identified graph, selecting at least one cell corresponding to the position from the to-be-identified image and the template image respectively, and confirming at least four vertexes located on an outer frame from the selected cells respectively, the confirmed vertexes in the to-be-identified image corresponding to the confirmed vertexes in the template image to form at least four groups of corresponding vertex coordinates; calculating a perspective transformation matrix by using the corresponding vertex coordinates. The method comprises the steps that a blank picture is created, the size of the blank picture is the same as that of a template picture, a perspective transformation matrix is used for calculating transformed pixel values of all pixels in a to-be-recognized picture, the pixel values are written into the blank picture, and therefore a perspective transformation image is obtained, and image normalization is achieved. According to the invention, the problem of difference between the bill image and the standard sample ticket is solved, and the bill image to be identified can be converted into a format similar to that of the standard sample ticket.

Description

technical field [0001] The invention relates to a method and equipment for normalizing images based on table features, and belongs to the field of bill automatic recognition. Background technique [0002] For notes typesetting in a fixed format, the printed paper notes will also have a fixed format, please refer to figure 1 , such as special value-added tax invoices, all have fixed areas such as "buyer", "password area", "seller", and "remarks". Therefore, in the process of identifying the bill, it is necessary to identify the relatively fixed overprinting area. [0003] Using templates to automatically identify bills is to design a set of reusable templates based on bills with a fixed format. The position of each area to be recognized and the width and height of the area are specifically marked in the template. According to the position in the template, the electronic version of the bill obtained after scanning or shooting is cut to obtain an image of the precise area, an...

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): G06F17/24G06F17/25G06T3/40G06T15/20G06F40/189
Inventor 吴建杭陈文传杜保发方恒凯郝占龙林玉玲庄国金
Owner 厦门商集网络科技有限责任公司
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