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

Method and device for deep learning and identification of road conditions and climate

A technology of deep learning and recognition methods, applied in character and pattern recognition, electrical components, instruments, etc., can solve problems such as inability to distinguish road conditions and climate information, and achieve the effect of reducing data processing pressure

Active Publication Date: 2018-06-19
DALIAN ROILAND SCI & TECH CO LTD
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in the current car assisted driving or active driving or even unmanned driving, the road condition and climate information (rainy day road condition, snowy day road condition, sunny day road condition, night road condition, etc.) There are certain risks in driving or active driving

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
  • Method and device for deep learning and identification of road conditions and climate
  • Method and device for deep learning and identification of road conditions and climate

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0021] Embodiment 1: a kind of road condition climate deep learning and identification method, comprising: S1. vehicle-mounted visual sensor collects road condition climate scene image, and image data is sent to cloud server; S2. cloud server distributes the road condition climate scene image data that receives to CNNs learns the model, performs distributed parallel computing, and trains to obtain a deep image feature classifier. Since the amount of data transmitted from the vehicle-mounted visual sensor to the cloud server is very large, it is difficult for ordinary servers to have such computing capabilities, so many servers need to be combined. That is to carry out distributed parallel computing, so as to solve the problem that the amount of data is too large and difficult to calculate. S3. The depth image feature classifier is sent to the vehicle recognition terminal to identify the road condition climate scene reflected by the road condition climate scene image newly colle...

Embodiment 2

[0023] Embodiment 2: have the technical scheme identical with embodiment 1, more specifically: described method also comprises: S4. vehicle-mounted recognition terminal scene recognition fails, then the road condition climate scene image data that vehicle-mounted vision sensor collects currently is marked and It is transmitted to the cloud server, and the cloud server distributes the received road conditions and climate scene image data to the CNNs learning model, and trains a new depth image feature classifier, which is sent to the vehicle recognition terminal and replaces the previous depth image feature classifier.

[0024] This technical solution enables learning and recognition to be in an alternate and uninterrupted state, updates the in-depth training classifier in real time, gradually strengthens the number and ability of road conditions that the classifier can recognize, and uses massive vehicle road condition image collection as a source of continuous learning, real-t...

Embodiment 3

[0028] Embodiment 3: have the same technical scheme as embodiment 1 or 2, more specifically: the vehicle identification terminal recognizes the current road condition climate, then identification information is sent to the control terminal, and the current road condition climate of the car owner is prompted by the control terminal. A preferred solution is: the prompt can be any of voice prompts, screen display prompts or a combination thereof. The realization of this kind of technical scheme makes it possible to give reminders to the driver, so that the driver can actively grasp the real-time weather of the road conditions during the current driving process, and specify corresponding driving arrangements and strategies.

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 road condition and climate deep learning and recognition method and device belong to the field of road condition and climate recognition, and are used to solve the problem of rapid and high-precision recognition of road conditions and climate by vehicles. The image data is sent to the cloud server; the cloud server distributes the received road conditions and climate scene image data to the CNNs learning model, and the training obtains the depth image feature classifier; the depth image feature classifier is sent to the vehicle recognition terminal to identify the vehicle vision The road condition climate scene reflected by the road condition climate scene image newly collected by the sensor. The effect is: it can realize fast and high-precision recognition effects of cloud learning and terminal recognition.

Description

technical field [0001] The invention belongs to the field of road condition climate recognition and relates to a method and device for road condition climate recognition. Background technique [0002] In recent years, 4G Internet of Vehicles has attracted the attention of many automobile companies and technology companies, and many scientific research institutions and enterprise R&D centers have invested a lot of energy in research and development of related products. Especially in terms of assisted driving and in-vehicle entertainment, it has attracted a large number of start-up companies and researchers. At present, relying on 4G Internet of Vehicles, the information collected by the vehicle sensor can be transmitted to the remote server for processing, which reduces the processing pressure of the vehicle terminal, and at the same time can obtain other useful information from the server to the vehicle terminal in real time to help drivers. Enjoy driving more safely and co...

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): G08G1/01G06K9/62H04L29/08
CPCH04L67/12G08G1/01G06F18/241G06F18/214
Inventor 田雨农吴子章周秀田于维双陆振波
Owner DALIAN ROILAND SCI & TECH CO LTD
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