Text detection method and device, electronic equipment and storage medium

A text detection and text technology, applied in character and pattern recognition, instruments, calculation models, etc., can solve problems such as high time consumption and affect OCR time consumption, and achieve the effect of reducing time consumption and solving the overall high delay.

Active Publication Date: 2020-09-15
GUANGDONG XIAOTIANCAI TECH CO LTD
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, it is necessary to find the contour for each text line area mask, and for the input image with a relatively dense text line area, the overall time-consuming to find the contour is high, reaching more than 400ms, accounting for 80% of the overall text line detection algorithm. %-90%, so it affects the overall time-consuming of OCR

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
  • Text detection method and device, electronic equipment and storage medium
  • Text detection method and device, electronic equipment and storage medium
  • Text detection method and device, electronic equipment and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0076] see figure 1 , figure 1 It is a schematic flowchart of a text detection method disclosed in an embodiment of the present invention. Such as figure 1 As shown, the text detection method includes the following steps:

[0077] 110. Acquire a mask map of the text line area mask of the target picture, where the size of the mask map is the same as that of the target picture.

[0078] The target image is an image input by the user. The image input by the user may be an image obtained by taking a picture of the document through an image acquisition device, or an image downloaded by the user from the Internet, which is not limited here. There is one or more text lines in the target image, and there is no limitation on whether the text lines are horizontal or not.

[0079] There can be multiple ways to obtain the text line area mask of the target picture. In the embodiment of the present invention, a text line detection network model based on deep learning is used to realize ...

Embodiment 2

[0099] see Figure 5 , Figure 5 is a schematic flowchart of another text detection method disclosed in the embodiment of the present invention. Such as Figure 5 As shown, the text detection method includes the following steps:

[0100] 310. Acquire a mask map of the text line area mask of the target picture, where the size of the mask map is the same as that of the target picture.

[0101] 320. Determine the value of each pixel in the mask map. In the text line area mask numbered i, the value of each pixel in the text line area mask is i, and the text line area in the mask map The values ​​of the remaining pixels outside the mask are 0; 1≤i≤M, where M is the total number of text line area masks corresponding to the target image.

[0102] 330. Subtract the value of the pixel point in the jth row of the mask image from the value of the corresponding pixel point in the j+1th row to obtain a new value of the pixel point in the jth row or j+1th row, where 1≤j≤N , N is the to...

Embodiment 3

[0117] see Figure 6 , Figure 6 It is a schematic flowchart of another text detection method disclosed in the embodiment of the present invention. Such as Figure 6 As shown, the text detection method includes the following steps:

[0118] 410. Acquire a mask map of the text line area mask of the target picture, where the size of the mask map is the same as that of the target picture.

[0119] 420. Determine the value of each pixel in the mask map. In the text line area mask numbered i, the value of each pixel in the text line area mask is i, and the text line area in the mask map The values ​​of the remaining pixels outside the mask are 0; 1≤i≤M, where M is the total number of text line area masks corresponding to the target image.

[0120] 430. Subtract the value of the pixel point in the jth row of the mask image from the value of the corresponding pixel point in the j+1th row to obtain the new value of the pixel point in the jth row or j+1th row, where 1≤j≤N , N is t...

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 text detection method and device, electronic equipment and a storage medium. The method comprises the steps of obtaining a mask graph of a text line regionmask of a target picture; determining the value of each pixel point in the mask map, and in the text line area mask with the number of i, setting the value of each pixel point in the text line area mask as i; subtracting the value of the pixel point corresponding to the (j+1)th row from the value of the pixel point in the jth row in the mask image to obtain a new value of the pixel point in the jth row or the (j+1)th row; respectively forming first boundary information and second boundary information corresponding to the text line area mask with the number of i by a set of pixel points equal to -i and i in the new value; and utilizing the first boundary information and the second boundary information to construct a text line contour corresponding to the text line region mask with the number of i. By implementing the embodiment of the invention, the outline of each text line can be quickly determined, and the time consumption of the whole text recognition is reduced.

Description

technical field [0001] The present invention relates to the technical field of text detection, in particular to a text detection method, device, electronic equipment and storage medium. Background technique [0002] In text recognition technology, photographed images are greatly affected by the environment. In text recognition, it is necessary to detect text lines to obtain the best bounding box of text lines, so as to recognize the text in the bounding box. [0003] The existing classic text detection technology is mainly based on the PSENet text line detection algorithm, which combines FPN and PSE technology, first detects each text line through FPN, and then performs post-processing based on PSE, that is, the progressive scale expansion algorithm Finally, the output is a multi-classified mask map for the text area and background, that is, a matrix with only one channel of the same size as the input image is output, and each value is 0, 1, 2...n( n is the number of comple...

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): G06K9/20G06K9/34G06K9/62G06N20/00
CPCG06N20/00G06V10/22G06V10/267G06F18/24
Inventor 尹磊邓小兵张春雨
Owner GUANGDONG XIAOTIANCAI TECH CO LTD
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