Text positioning box correction method and system based on convolutional neural network

A convolutional neural network and text positioning technology, which is applied in biological neural network models, neural architectures, instruments, etc., can solve the problems of unframed text, insufficient precision, and text included, so as to reduce uniform size and reduce the amount of calculation. , targeted effect

Pending Publication Date: 2020-10-16
厦门商集网络科技有限责任公司 +1
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] In the prior art, when some text positioning methods are used for text positioning of text pictures, the phenomenon of blurred and inaccurate text positioning boundaries will appear. Most of this phenomenon is that the text positioning box cannot completely locate all the text; The anchor box is too large and contains irrelevant text
Although these positioning methods can locate most of the required text, the positioning at the corners of the four vertices of the text area still ha

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 positioning box correction method and system based on convolutional neural network
  • Text positioning box correction method and system based on convolutional neural network
  • Text positioning box correction method and system based on convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0062] Example one

[0063] Such as figure 1 As shown, the text positioning frame correction method based on convolutional neural network includes the following steps:

[0064] S1: Obtain multiple text images to be located.

[0065] S2: Input a plurality of acquired text images to be positioned into a text detection model, the text detection model performs coarse text positioning on the text image to be positioned, outputs the positioned text image, and a text positioning frame to be corrected The coordinates of the upper and lower ends of the left and right ends.

[0066] Such as figure 2 As shown, this picture is a text image, that is, an image with text. The rectangular box is the text positioning box obtained by the text detection model after rough positioning. It can be seen that the upper part of this text positioning box does not include the upper part of all the text, which will cause subsequent recognition errors. The corner points marked by the four dots are the four poin...

Example Embodiment

[0114] Example two

[0115] A text positioning frame correction system based on a convolutional neural network includes a memory and a processor. The memory stores instructions, and the instructions are adapted to be loaded by the processor and execute the following steps:

[0116] S1: Obtain multiple text images to be located.

[0117] S2: Input a plurality of acquired text images to be positioned into a text detection model, the text detection model performs coarse text positioning on the text image to be positioned, outputs the positioned text image, and a text positioning frame to be corrected The coordinates of the upper and lower ends of the left and right ends.

[0118] S3: Establish a text positioning frame correction model, and train the text positioning frame correction model.

[0119] S4: After cropping and scaling the text positioning box to be corrected and its corresponding image content, input the trained text positioning box correction model, output the correction value...

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 a text positioning box correction method and system based on a convolutional neural network, and the method comprises the steps: building a neural network for text positioningbox correction, carrying out the independent training of the left and right ends of a text positioning box, setting a subsequent verification step for verifying a result, and guaranteeing that a trained model meets the requirements. Compared with the prior art, the text box correction model obtained by training is used for correcting the text positioning box in the text picture, so that the positioning precision of the text detection method can be effectively improved, a more accurate text positioning box is obtained, and the accuracy of picture text recognition and the practicability of a text detection and recognition system are improved.

Description

technical field [0001] The invention relates to a text positioning frame correction method and system based on a convolutional neural network, and belongs to the field of OCR text recognition. Background technique [0002] OCR (optical character recognition) character recognition refers to the process in which electronic devices (such as scanners or digital cameras) check the characters printed on paper, and then use character recognition methods to translate the shape into computer text; that is, the text data is scanned, and then The process of analyzing and processing image files to obtain text and layout information. With the advancement of image processing technology in recent years, there is an increasing demand for OCR text recognition. In the OCR text recognition process, the most basic step is to accurately locate the text in the image. Only by completing accurate positioning can the subsequent recognized text be correct and without omissions. At present, the two ...

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
IPC IPC(8): G06K9/20G06K9/32G06K9/62G06N3/04
CPCG06V10/24G06V10/22G06V30/10G06N3/045G06F18/214
Inventor 茹超飞黄征
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