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

Detection identification and detection identification network training methods and apparatuses, device and medium

A technology for identifying networks and identifying methods, which is applied in the field of detection and identification and detection and identification networks, and can solve problems such as low resolution, complex and difficult problems

Active Publication Date: 2018-06-29
BEIJING SENSETIME TECH DEV CO LTD
View PDF6 Cites 30 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] However, text detection and recognition in natural scenes is very challenging. Factors such as different background scenes, low resolution, different fonts, different lighting conditions, different sizes, different tilt directions, and blur all make this problem become very complex and difficult

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
  • Detection identification and detection identification network training methods and apparatuses, device and medium
  • Detection identification and detection identification network training methods and apparatuses, device and medium
  • Detection identification and detection identification network training methods and apparatuses, device and medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0118] Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that the relative arrangements of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present invention unless specifically stated otherwise.

[0119] At the same time, it should be understood that, for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0120] The following description of at least one exemplary embodiment is merely illustrative in nature and in no way taken as limiting the invention, its application or uses.

[0121] Techniques, methods and devices known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, such techniques, methods and devices should be considered part of the descript...

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

Embodiments of the invention disclose detection identification and detection identification network training methods and apparatuses, a device and a medium. The detection identification method comprises the steps of inputting a to-be-processed image to a detection identification network, wherein the detection identification network comprises a shared network layer, a detection network layer and anidentification network layer; outputting shared layer features of the to-be-processed image through the shared network layer; inputting the shared layer features to the detection network layer, outputting detection layer features of the to-be-processed image through the detection network layer, and based on the detection layer features, obtaining text box information comprising texts in the to-be-processed image; and inputting the shared layer features and the text box information to the identification network layer, and outputting text contents in a text box through the identification network layer. Therefore, the repeated feature extraction of the image is reduced, the processing efficiency is improved, and the text detection and identification efficiency and speed are improved.

Description

technical field [0001] The invention relates to computer vision technology, in particular to a training method, device, equipment and medium for detection and recognition and detection and recognition networks. Background technique [0002] Text detection and recognition in natural scenes is an important problem in the field of image understanding and image restoration. Accurate text detection and recognition can be used for many problems, such as image search under large data sets, automatic translation, guidance for the blind, robot navigation, etc. [0003] However, text detection and recognition in natural scenes is very challenging. Factors such as different background scenes, low resolution, different fonts, different lighting conditions, different sizes, different tilt directions, and blur all make this problem become very complex and difficult. Contents of the invention [0004] An embodiment of the present invention provides a technical solution for character re...

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): G06K9/00G06K9/34G06K9/62
CPCG06V30/413G06V10/267G06V30/10G06F18/214
Inventor 刘学博梁鼎
Owner BEIJING SENSETIME TECH DEV CO LTD