A Single-line Text Recognition Method Based on Pointing Gesture

A text recognition and gesture technology, applied in the field of computer vision, can solve problems such as deformation, bending, reordering, and the inability to determine where the user wants to read

Active Publication Date: 2021-07-06
杭州视氪科技有限公司
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] There are many text areas in natural scene images, and it is impossible to determine where the user wants to read
There is deformation and bending in the single-line text area in the natural scene image, and the traditional optical character recognition algorithm will have text sorting errors
It takes a long time to directly apply neural network algorithms to process natural scene images, and there is still a problem of reordering after single character recognition

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 Single-line Text Recognition Method Based on Pointing Gesture
  • A Single-line Text Recognition Method Based on Pointing Gesture
  • A Single-line Text Recognition Method Based on Pointing Gesture

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0022] In order to understand the purpose, features and advantages of the present invention more clearly, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0023] figure 1 It is a flowchart of an embodiment of a single-line text image recognition method based on pointing gestures in the present invention. The method includes:

[0024] Step 1: Acquire text images of natural scenes with pointing gestures.

[0025] Through the camera on the visual aid glasses worn by the user, the text color image of the user pointing to the text with the index finger is obtained, such as figure 2 shown.

[0026] Step 2: The pointing gesture is recognized.

[0027] Collect images of pointing gestures as training samples, use machine learning methods to train pointing gesture classifiers, use the classifier to identify pointing gestures in text color images, and obtain gesture positions handRect and gestur...

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 single-line character recognition method based on pointing gestures. By detecting the user's pointing gestures and obtaining the pointing position, the upper character image is segmented, and the tilt correction is realized, and the curved character line is segmented to obtain a complete line of text area images, and then A neural network-based text recognition is performed on images of single-line text regions. The invention can acquire the single-line characters above the pointing position of the finger, reduces the time-consuming recognition, and can realize the segmentation of curved character lines.

Description

technical field [0001] The invention relates to the field of computer vision, in particular to a single-line character recognition method based on pointing gestures. Background technique [0002] With the development of the field of computer vision, the visual aid glasses industry has gathered more scientific and technological achievements. The camera added on the visual aid glasses can obtain more image information. Through the text image information, it can provide users with visual impairments to read the text content. However, if the hand-held text is photographed in a natural scene with an auxiliary device, the image has text deformation, page rotation, uneven illumination, and multiple scales. In the existing technical solutions, the traditional optical character recognition algorithm is mainly oriented to neatly typed high-quality text images, or the application of neural network algorithms for natural scene image text recognition. In view of the situation that onl...

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 & AuthorityPatents(China)
IPC IPC(8): G06K9/00G06K9/32G06K9/34
CPCG06V40/11G06V40/113G06V30/40G06V10/243G06V10/267
Inventor严江江林龙庆于红雷孙俊伟
Owner杭州视氪科技有限公司