Image data target detection rectangular frame automatic labeling method and system

A target detection and image data technology, applied in the field of image processing, can solve problems such as time-consuming and cumbersome processes

Pending Publication Date: 2021-08-31
NAT INNOVATION INST OF DEFENSE TECH PLA ACAD OF MILITARY SCI
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In order to overcome the above-mentioned defects of the prior art, the object of the present invention is to

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
  • Image data target detection rectangular frame automatic labeling method and system
  • Image data target detection rectangular frame automatic labeling method and system
  • Image data target detection rectangular frame automatic labeling method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0079] The image data target detection rectangular frame automatic labeling method provided by Embodiment 1 of the present invention (that is, the generation method of the target detection rectangular frame), such as figure 1 As shown, it specifically includes the following steps:

[0080] S11. Collect the image of the target object, and obtain the length, width and height of the target object, as well as the world coordinates of the center point of the target object, the posture information of the target object, the world coordinates of the image collector, and the posture information of the image collector;

[0081] S12. According to the length, width and height of the target, as well as the world coordinates of the center point of the target, the attitude information of the target, the world coordinates of the image collector and the attitude information of the image collector, obtain the minimum circumscribed cuboid of the target through coordinate transformation The pixel...

Embodiment 2

[0216] The embodiment of the present invention is based on the world coordinates of the center point of the target in S1422 in the first embodiment, the world coordinates of the image collector and the attitude information of the image collector, and sequentially pass through the world coordinate system to the coordinate system of the image collector, and the image collector The coordinate transformation from the coordinate system to the image coordinate system, and from the image coordinate system to the pixel coordinate system, to obtain the pixel coordinates of the center point of the target object in the pixel coordinate system, is described in detail as follows:

[0217] S14221', according to the world coordinate P of the center point of the target w =[x w the y w z w ] T , the world coordinates of the camera O c =[x o the y o z o ] T and attitude information (including yaw angle θ, pitch angle and roll angle ω), calculate the coordinates P of the target in the ...

Embodiment 3

[0245] Embodiment 3 of the present invention also provides a method for automatically marking a rectangular frame of image data target detection, such as Figure 6 As shown, it specifically includes the following steps:

[0246] S31. Generate a rectangular frame for target detection in the image;

[0247] S32. Obtain the category of the target object in the image, and add a category label to the target detection rectangle in the image according to the category of the target object, so as to obtain final labeling data of the target object in the image.

[0248] Wherein, in S31, any method in the first or second embodiment may be used to generate a rectangular frame for object detection in the image.

[0249] Based on the same inventive concept, Embodiment 3 of the present invention also proposes an automatic labeling system for image data object detection rectangular frames, which specifically includes the following components:

[0250] Generating means 41, for generating the...

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 an image data target detection rectangular frame automatic labeling method and system. The labeling method comprises the steps of collecting a target object image and related information of a target object and an image collector; calculating pixel coordinates of each vertex of the minimum circumscribed cuboid of the target object; constructing a minimum rectangular bounding box according to the pixel coordinates of the vertexes of the minimum bounding cuboid of the target object, and determining the pixel coordinates of the reference vertexes of the minimum rectangular bounding box; and generating a target detection rectangular frame based on the pixel coordinates of the reference vertexes of the minimum rectangular bounding box, or generating the target detection rectangular frame based on the pixel coordinates of the reference vertexes of the minimum rectangular bounding box and the pixel coordinates of the central point of the target object. According to the method and the system, the automatic generation of the target detection rectangular frame can be really realized on the premise of not needing human intervention; besides, the operation process can be simplified, and the accuracy of labeling is ensured.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to an automatic labeling method and system for a rectangular frame of image data target detection. Background technique [0002] In the existing technology, supervised learning requires a large amount of labeled data, which can be completely labeled manually, automatically labeled by a machine, or a mixture of humans and machines. These three modes are called manual labeling, automatic labeling and semi-automatic labeling. For example, in automatic driving, the perception system has a great demand for labeling, especially obstacle perception. Its automatic labeling can mainly use the mutual labeling between different sensors. There are three mainstream sensors for obstacle perception: LiDAR, Camera, and Radar. From the perspective of the data flow of automatic labeling, it is generally to rely on lidar and millimeter-wave radar to label the camera. This autom...

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): G06T7/62G06T7/66G06T7/70G06K9/62
CPCG06T7/62G06T7/66G06T7/70G06T2207/30204G06F18/214
Inventor 苏龙飞王世雄易晓东王之元凡遵林沈天龙黄强娟
Owner NAT INNOVATION INST OF DEFENSE TECH PLA ACAD OF MILITARY SCI
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