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

Picture character recognition method and device

A text recognition and picture technology, applied in character recognition, digital ink recognition, character and pattern recognition, etc., can solve problems such as difficult feature extraction, multi-word adhesion cannot be accurately segmented, large differences in writing, etc., to achieve accuracy and completeness sexual effect

Active Publication Date: 2021-02-05
TAIJI COMP
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In view of the characteristics of handwritten text: large differences in writing, difficult to extract features, and inability to accurately segment multiple words, etc., how to realize the cutting of handwritten text images and improve the accuracy and integrity of handwritten handwriting has become a technical problem that needs to be solved urgently

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
  • Picture character recognition method and device
  • Picture character recognition method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0027] Aiming at the characteristics of handwritten text: large differences in writing, difficult to extract features, and inability to accurately segment multiple characters, we study data preprocessing techniques for enhancing the quality of handwritten text, and realize image rotation correction; text line tilt correction; handwriting Width normalization, and the removal of other common noise and interference, etc., while effectively removing image noise and pollution interference, improve the accuracy and integrity of handwriting.

[0028] In addition, the number of Chinese categories is huge, and the method based on deep learning requires a large number of training samples. If all of them are obtained through manual labeling, it will consume a lot of manpower and time, and there is a problem that the coverage of real sample categories is small, so the trained model will generalize when it is actually used. poor ability. Based on this, automatic sample synthesis is propose...

Embodiment 2

[0045] A picture and text recognition device, the device includes a processor and a memory, and a computer program is stored in the memory, and the computer program is executed by the processor to implement the method in Embodiment 1.

[0046] Those skilled in the art can understand that all or part of the processes in the methods of the above-mentioned embodiments can be completed by instructing related hardware through computer programs, and the programs can be stored in a computer-readable storage medium. During execution, it may include the processes of the embodiments of the above-mentioned methods. Wherein, the storage medium may be a magnetic disk, an optical disk, a read-only memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a flash memory (Flash Memory), a hard disk (Hard Disk Drive, abbreviation: HDD) or a solid-state drive (Solid-State Drive, SSD), etc.; the storage medium may also include a combination of the above-mentioned types ...

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 discloses a picture character recognition method and device, and the method comprises the steps: segmenting out different window pictures from a to-be-processed image through employing aselected picture of an irregular sliding window and not depending on a preset small frame, enabling the window pictures to serve as the input of a neural network, carrying out the classification andrecognition of the pictures through the neural network, and then carrying out the decoding for character recognition. And the font identification accuracy and integrity are improved.

Description

technical field [0001] The invention relates to the field of artificial intelligence, in particular to a method and device for recognizing pictures and characters. Background technique [0002] With the development of emerging technologies such as sensor technology, intelligent computing technology, cloud computing, and big data, traditional interactive methods such as mouse and keyboard can no longer meet the needs of new, efficient, and convenient human-computer interaction. The natural and efficient interaction between humans and systems, Accurate understanding of user intent and other key issues urgently need to be solved. The application realization of human-computer interaction collaboration can make the system operation more flexible and humanized. thus receiving wider attention. [0003] In the human-computer interaction environment, there are two different text interaction methods: one is text interaction through keyboard input, and the other is handwritten text i...

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): G06K9/00G06K9/34G06K9/62G06N3/04G06N3/08
CPCG06N3/084G06V30/333G06V30/36G06V30/153G06V30/10G06N3/047G06N3/045G06F18/2415G06F18/214
Inventor 黄海峰
Owner TAIJI COMP