Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Data enhancement method and system based on digital twinning and image conversion

An image conversion and data technology, applied in image enhancement, image data processing, graphic image conversion and other directions, can solve the problems of inability to generate, uneven distribution of vehicles, low degree of freedom, etc., to achieve the effect of high generation efficiency and improved recognition rate

Pending Publication Date: 2022-02-15
SHANGHAI JIAO TONG UNIV
View PDF1 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] Most of the above examples are based on the conversion of real pictures from day to night. There are no label information, uneven distribution of vehicles, low degrees of freedom, etc., and it is impossible to generate sufficient, accurate, and rich data sets to supply the recognition nerves. network to learn

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
  • Data enhancement method and system based on digital twinning and image conversion
  • Data enhancement method and system based on digital twinning and image conversion
  • Data enhancement method and system based on digital twinning and image conversion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0072] According to a data enhancement method based on digital twin and image conversion provided by the present invention, such as figure 1 shown, including:

[0073]Step S1: Obtain high-definition video data through the road test camera, and use the target recognition algorithm to identify the 2D coordinates, vehicle type, and color data of the vehicle in the video, based on the fixed camera angle of view and road height, for each camera A conversion function F(Xreal, Yreal)=(Xvirtual, Yvirtual, 0), wherein, Xreal, Yreal represents the position of the vehicle recognized from each frame of the video, and for a 1080P video stream, the range of Xreal is [0,1920 ], the range of Yreal is [0,1080], Xvirtual, Yvirtual means that in the corresponding city scene, the vehicle’s virtual world X, Y coordinates, Z coordinates are 0 by default; then the coordinates of the vehicle in the virtual world are obtained through the mapping function and Perform 3D reconstruction;

[0074] Step ...

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 provides a data enhancement method and system based on digital twinning and image conversion. The method comprises the steps: S1, obtaining video data through the shooting of a road test camera, recognizing the 2D coordinate, the vehicle type and the color data of a vehicle in a video through a target recognition algorithm according to the obtained video data, and carrying out the recognition of the 2D coordinate, the vehicle type and the color data; coordinates of the vehicle in the virtual world are obtained through a mapping function, and three-dimensional reconstruction is carried out; s2, calculating according to the coordinates of the vehicle in the virtual world and the coordinates of the camera to obtain the screen coordinates of the bounding box; s3, utilizing a weather system constructed by a particle system and shader, and randomly replacing the vehicle type and color of the vehicle to generate a virtual picture set; and S4, generating a content code c and a style code s from the virtual data set through a GAN network, and converting the picture into another style by combining different style codes.

Description

technical field [0001] The present invention relates to the field of computer vision and image conversion, in particular, to a data enhancement method and system based on digital twin and image conversion, and more specifically, to converting a virtual traffic scene picture generated by a digital twin system into a real style picture to perform Algorithms that harden datasets. Background technique [0002] The present invention uses the digital twin traffic as the basic data source, and can generate a virtual scene corresponding to the real scene through the digital twin technology, and reproduce the vehicle in three dimensions. By changing the model, color, weather and other attributes and combining them, the generated virtual picture can cover almost all extreme environments and vehicles of different subdivisions in the real scene. Then through the image conversion method, the virtual scene picture is converted into a real photo style picture, so as to obtain sufficient e...

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): G06T15/02G06T3/00G06T7/73G06N3/04G06N3/08G06V10/774
CPCG06T15/02G06T7/73G06N3/08G06T2207/20081G06T2207/30252G06N3/045G06F18/214G06T3/08
Inventor 杨旭波胡泽宇
Owner SHANGHAI JIAO TONG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com