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

Multi-sensor target tracking method for robots based on msr-cnn

A target tracking, multi-sensor technology, applied in the field of intelligent robots, can solve problems such as insufficient real-time robustness and restricting the development of SiamR-CNN

Active Publication Date: 2021-04-16
江苏德劭信息科技有限公司
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, Siam R-CNN technology also has many shortcomings, such as insufficient real-time robustness, etc., which greatly limit the development of Siam R-CNN in the field of target tracking technology.

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
  • Multi-sensor target tracking method for robots based on msr-cnn
  • Multi-sensor target tracking method for robots based on msr-cnn
  • Multi-sensor target tracking method for robots based on msr-cnn

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0031] The present invention will be described in further detail below in conjunction with the accompanying drawings and specific embodiments:

[0032] The invention proposes a multi-sensor target tracking method for a robot based on MSR-CNN, uses multi-scale data images for target recognition through corresponding algorithms, and processes collected images from various data sources, so that the target can be tracked efficiently and accurately. like figure 1 is the system architecture diagram.

[0033] First, the robot uses a binocular camera to collect multi-angle target images. Image preprocessing mainly performs multi-scale downsampling on the multi-angle data collected by the binocular camera to establish a multi-scale data set, and then performs saliency detection on the multi-scale data set. , to obtain a multi-scale saliency feature set.

[0034] The scales of image downsampling are ×1, ×4, and ×16, respectively, where ×1 represents the original image, × represents th...

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 present invention designs a robot multi-sensor target tracking method based on MSR-CNN. The target tracking technology has been widely used in many fields such as intelligent monitoring and unmanned driving, but the target tracking algorithm still faces huge challenges in complex environments. Aiming at this problem, the present invention proposes a robot multi-sensor target tracking method based on MSR-CNN. The robot uses the binocular camera to collect images, and utilizes the multi-angle information collected by the robot. Multi-scale decomposition of images for image saliency detection to segment the key positions of the image. Finally, MSR-CNN is used to identify the image target to achieve the purpose of robot target tracking.

Description

technical field [0001] The invention relates to an intelligent robot, and particularly designs a robot multi-sensor target tracking method based on MSR-CNN. Background technique [0002] The target tracking technology has been widely used in many fields such as intelligent monitoring and unmanned driving, but the target tracking algorithm still faces huge challenges in the complex environment. Especially in the field of unmanned driving, target tracking calculation is particularly important, and misjudgment will cause huge damage to the user's personal safety and social economy. [0003] In recent years, with the continuous progress of deep learning technology, the target tracking technology based on deep learning has been greatly developed. The Siam R-CNN network adopts a small-sized convolution kernel, which is widely used in face recognition, target tracking technology, etc. However, Siam R-CNN technology also has many shortcomings, such as insufficient real-time robust...

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 Patents(China)
IPC IPC(8): G06K9/62G06K9/46G06F17/14
CPCG06F17/141G06V10/464G06F18/214
Inventor 邓杨敏
Owner 江苏德劭信息科技有限公司
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