Multi-module collaborative object recognition system and method based on deep learning

An object recognition and deep learning technology, applied in the field of deep learning, can solve the problems that the 4G network bandwidth cannot meet the real-time transmission of high-quality video images, and cannot guarantee the real-time performance of object recognition, so as to shorten the image processing time, avoid time delay, reduce The effect of CPU usage

Active Publication Date: 2021-04-27
山东奥邦交通设施工程有限公司
View PDF22 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] However, the inventors found that with the increase of video image quality, the existing 5G network infrastructure does not cover all, and the 4G network bandwidth cannot meet the real real-time transmission of high-quality video images. The data is further processed by computer vision for object recognition, and the real-time performance of object recognition cannot be guaranteed.

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-module collaborative object recognition system and method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0052]Referfigure 1 Introvertial learning-based multi-module synergistic object identification system, including integrated and synergistic video input modules, video processing subsystem modules, smart video engine modules, neural network acceleration engine modules, video graphic sub-system modules And video output modules.

[0053]In the specific implementation, in order to ensure the accuracy of the post-data processing, the video input module, the video processing subsystem module, the smart video engine module, the neural network acceleration engine module, the video graphic system module, and the video output module are received after receiving the start command. Start and initialize the initialization operation.

[0054]During the initialization operation, the initialization of the neural network acceleration engine module includes loading a well-trained neural network model that has been trained in a particular format. Before loading, it is necessary to format the neural network ...

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 belongs to the technical field of deep learning, and provides a multi-module collaborative object recognition system and method based on deep learning. The system comprises a video input module, a video processing subsystem module, an intelligent video engine module, a neural network acceleration engine module, a video graphic subsystem module and a video output module which are integrated and work cooperatively. According to the invention, real-time identification of an object is realized by using multi-module cooperation, the problem that a camera image needs to be uploaded to a server for identification processing at present is solved, time delay caused by network delay or limitation of network bandwidth is avoided, and real-time identification is realized.

Description

Technical field[0001]The present invention belongs to the field of depth learning, and more particularly to a depth learning multimode synergistic object identification system and method.Background technique[0002]The statement of this section is merely the background technology information associated with the present invention, which is not necessarily constituted in prior art.[0003]Visual information processing is based on external sensing data to construct an intelligent system that simulates human visual capabilities and judges and identifies the target, wherein the object identifies the basis of visual information processing technology. With the popularity of the computer and the intelligent terminal and the rapid development of the Internet, the rapid extension of the field video large data application field has challenged object identification technology. The current object identification technology should have the characteristics of high efficiency, high performance and even ...

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): G06K9/62G06N3/04G06N3/08
CPCG06N3/04G06N3/08G06F18/241
Inventor 奚照明杨哲邵强梁昭蔡达张辉马琳
Owner 山东奥邦交通设施工程有限公司
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