target recognition and scene modeling system based on RGBD

A target recognition and scene modeling technology, applied in scene recognition, character and pattern recognition, instruments, etc., can solve the problems of poor target recognition and scene modeling accuracy, data without reference and comparison, slow data processing speed, etc. The extraction and processing rules are consistent, the data processing speed is improved, and the judgment is accurate.

Pending Publication Date: 2019-02-15
BENEWAKE BEIJING TECH CO LTD
View PDF3 Cites 14 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the embodiments of the present invention is to propose a RGBD-based target recognition and scene modeling system, which is performed sequentially for target recognition and scene modeling in traditional systems. For the problem of poor scene modeling accuracy, the embodiment of the present invention uses target recognition and scene modeling to carry out simultaneously, and the data obtained by the target recognition module and the scene modeling module are exchanged, judged and fused to form a processing mechanism. The system not only Improves the speed of data processing and improves the accuracy of object recognition and scene modeling

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
  • target recognition and scene modeling system based on RGBD
  • target recognition and scene modeling system based on RGBD
  • target recognition and scene modeling system based on RGBD

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0054] Such as figure 1 As shown, a RGBD-based target recognition and scene modeling system includes a data acquisition module 1, a data processing module 2, a scene modeling algorithm module 3, and a target recognition algorithm module 4, and the data acquisition module 1 is used to collect RGBD Image, data acquisition module 1 selects network camera, radar or depth camera, and described RGBD image data includes RGB image and depth image, and RGB image is color image. Web cameras are used to capture color images, and radar or depth cameras are used to capture depth images.

[0055] The data processing module 2 is used to store local data sets, control the work of each functional module, and process the data transmitted from each functional module. The local data set stores training parameters, including: scene modeling parameters used to guide modeling after being trained by the local data set, such as classification method, geometric feature matching of scene objects, lengt...

Embodiment 2

[0089] Such as image 3 As shown, a RGBD-based target recognition and scene modeling system includes a data acquisition module 1, a data processing module 2, a scene modeling algorithm module 3, and a target recognition algorithm module 4, and the data acquisition module 1 is used to collect RGBD For images, the scene modeling algorithm module 3 is used for modeling matching and scene modeling; the target recognition algorithm module 4 is used for target detection matching and target recognition. The data acquisition module 1, the data processing module 2, the scene modeling algorithm module 3, and the object recognition algorithm module 4 are the same as those in the first embodiment.

[0090] In order to transmit RGBD data safely and conveniently, a prototype output port 6 is set between the data acquisition module 1 and the data processing module 2. The prototype output port 6 can choose a variety of communication ports, preferably a USB port.

[0091] In order to output i...

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 discloses a target recognition and scene modeling system based on RGBD, which relates to the technical field of target recognition. The system includes data acquisition module, data processing module, scene modeling algorithm module, target recognition algorithm module, data acquisition module is used to collect RGBD images; The data processing module is used for storing the local data set, controlling the work of each functional module, and processing the data transmitted from each functional module. The scene modeling algorithm module is used for scene modeling; A target recognition algorithm module is use for target recognition; Data exchange between scene modeling algorithm module and target recognition algorithm module, and work according to each other 's data. Through the simultaneous target recognition and scene modeling, the data obtained from the target recognition module and scene modeling module are exchanged, evaluated and fused to form a processing mechanism.The system solves the problem of slow processing speed in the traditional system, and improves the accuracy of target recognition.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to a laser radar target recognition and scene modeling method. Background technique [0002] Target identification technology is a technology that uses radar and computer to identify distant targets. Through the analysis of target characteristic information such as amplitude, phase, spectrum and polarization in radar echoes, various mathematical methods Multidimensional space transformation is used to estimate the size, shape, weight and physical characteristic parameters of the surface layer of the target, and finally, the identification judgment is made in the classifier according to the identification function determined by a large number of training samples. [0003] At present, when using lidar for target recognition, the point cloud of the object is generally segmented from the geometric features, that is, different objects are segmented through the point cloud...

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): G06K9/00G06K9/62
CPCG06V20/13G06F18/24G06F18/214
Inventor 疏达郑凯刘云浩李远
Owner BENEWAKE BEIJING 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