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

Method, device and equipment for decoding text line picture

A decoding method and image technology, applied in the field of image processing, can solve problems such as the discount of Transformer module recognition efficiency, achieve rapid recognition, overcome low decoding efficiency, and improve decoding efficiency

Pending Publication Date: 2021-08-13
BEIJING YOUZHUJU NETWORK TECH CO LTD
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The decoder in the Transformer model can only decode one character in the text line picture at a time. If there are many characters in the text line picture to be recognized, the recognition of the Transformer module will occur due to the high number of times the decoder needs to execute. Efficiency is greatly reduced

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
  • Method, device and equipment for decoding text line picture
  • Method, device and equipment for decoding text line picture
  • Method, device and equipment for decoding text line picture

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0046] In order to make the above objects, features and advantages of the present application more obvious and understandable, the embodiments of the present application will be further described in detail below in conjunction with the accompanying drawings and specific implementation methods. It can be understood that the specific embodiments described here are only used to explain the present application, but not to limit the present application. In addition, it should be noted that, for the convenience of description, only parts relevant to the present application are shown in the drawings, not all structures.

[0047] OCR technology currently has two main ideas: Connectionist Temporal Classification (English: ConnectionistTemporal Classification, referred to as: CTC) model and attention (English: attention) model, both of which can be used to recognize text information in pictures. Generally, the algorithm adopted by the CTC model may be a convolutional recurrent neural ne...

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 embodiment of the invention discloses a decoding method, device and equipment for a text line picture, a decoder of a Transform model is at least connected with a first module and a second module, when the decoder decodes the text line picture once, a previous decoding result is input into the decoder, the current decoding is performed on the text line picture, a first character is obtained from the first module and a second character is obtained from the second module; and the previous decoding result is spliced with the first character and the second character in sequence to obtain a current decoding result. Visibly, in the method provided by the invention, the Transform model decodes a plurality of characters at one time, so that the problem of low decoding efficiency caused by the fact that a decoder of the existing Transform model can only decode one character at one time is solved, the decoding efficiency of the text line picture is improved, and the recognition efficiency of the text line picture is improved.

Description

technical field [0001] The present application relates to the technical field of image processing, in particular to a decoding method, device and equipment for a text line picture. Background technique [0002] Optical Character Recognition (English: Optical Character Recognition, abbreviation: OCR) technology can recognize text information in pictures. Among them, the Transformer model, as an implementation of OCR technology, has a better recognition effect. [0003] The decoder in the Transformer model can only decode one character in the text line picture at a time. If there are many characters in the text line picture to be recognized, the recognition of the Transformer module will occur due to the high number of times the decoder needs to execute. The efficiency is greatly reduced. [0004] Based on this, it is urgent to provide a more efficient decoding method, which can quickly realize the decoding of characters in the text line picture, so as to improve the recogni...

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): G06K9/20G06F40/126
CPCG06F40/126G06V10/22G06V30/10
Inventor 蔡悦卢永晨黄灿王长虎
Owner BEIJING YOUZHUJU NETWORK TECH CO LTD