Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Vehicle collision detection method and device

A technology of vehicle collision and detection method, applied in the field of collision detection, can solve the problem of not being able to learn the deep semantic information of speech samples

Pending Publication Date: 2021-03-19
亚美智联数据科技有限公司
View PDF0 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] This application provides a vehicle collision detection method and device, which solves the technical problem that the existing technology can only learn the information of the voice sample in one-dimensional direction, but cannot learn the deep semantic information of the voice sample

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
  • Vehicle collision detection method and device
  • Vehicle collision detection method and device
  • Vehicle collision detection method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0057] figure 1 It is a method flowchart of an embodiment of a vehicle collision detection method of the present application, such as figure 1 as shown, figure 1 Including:

[0058] 101. Obtain an audio sample to be tested;

[0059] It should be noted that this application can obtain audio data collected from a vehicle-mounted recorder or other vehicle-mounted audio collection devices, and use the collected audio data as audio samples to be tested.

[0060] 102. Extract the MFCC feature vector of the audio sample to be tested;

[0061]It should be noted that the present application may use a common method for extracting MFCC feature vectors to extract the MFCC feature vectors of the audio samples to be tested.

[0062] In a specific embodiment, the MFCC feature extraction steps are as follows:

[0063] S1: Set the audio sample to be tested as x(n), and intercept the audio sample to be tested for 2s;

[0064] S2: After resampling the audio frequency of the 2s audio sample...

Embodiment 2

[0080] figure 2 It is a flow chart of a method for training a collision model in an embodiment of a vehicle collision detection method of the present application, such as figure 2 as shown, figure 2 Including:

[0081] 201. Acquire several second audio samples as a training set, where the second audio samples include collision audio samples and non-collision samples;

[0082] It should be noted that the present application may select a large number of collision audio samples and non-collision samples (ie, positive samples and negative samples) as a training set for training the collision model.

[0083] 202. Extract the MFCC feature vector of the second audio sample;

[0084] 203. Map the MFCC feature vector to a hwc three-dimensional feature vector;

[0085] 204. Input the hwc three-dimensional feature vector into the constructed collision model until the collision model converges to obtain a trained collision model.

[0086] The above steps 201-204 are steps for trai...

Embodiment 3

[0104] The above is the embodiment of the method of the present application, and the present application also provides an embodiment of a vehicle collision detection device, such as Figure 5 as shown, Figure 5 Including:

[0105] The first acquiring unit 401 is configured to acquire the audio sample to be tested;

[0106] The first extraction unit 402 is used to extract the MFCC feature vector of the audio sample to be tested;

[0107] The first mapping unit 403 is configured to map the MFCC feature vector to the hwc three-dimensional feature vector, so that the hwc three-dimensional feature vector conforms to the input format of the collision model;

[0108] A probability value acquisition unit 404, configured to input the hwc three-dimensional feature vector into the trained collision model to obtain a collision probability value;

[0109] The judging unit 405 is configured to determine that the audio sample to be tested is a collision audio sample if the collision probab...

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 vehicle collision detection method and apparatus. The method comprises the steps of obtaining a to-be-detected audio sample; extracting an MFCC feature vector of a to-be-tested audio sample; mapping the MFCC feature vector into an hwc three-dimensional feature vector, so that the hwc three-dimensional feature vector conforms to an input format of a collision model; inputting the hwc three-dimensional feature vector into a trained collision model to obtain a collision probability value; and if the collision probability value is greater than the preset probability value, determining that the to-be-tested audio sample is a collision audio sample. According to the invention, the technical problem that only the information of the voice sample in the one-dimensional direction can be learned and the deep semantic information of the voice sample cannot be learned in the prior art is solved.

Description

technical field [0001] The present application relates to the technical field of collision detection, in particular to a vehicle collision detection method and device. Background technique [0002] With the continuous improvement of the level of social development, the increase in the number of cars has brought a great burden to traffic, and the number of various traffic accidents is also increasing. This increase in the number increases the cost of traffic guidance. [0003] In this regard, the existing methods for vehicle collision detection have the following aspects: through image methods; through vehicle-mounted related sensors, such as infrared sensors, ultrasonic sensors, etc.; through audio methods. [0004] Based on the audio method, some existing methods mainly use the size of the audio to judge whether there is a collision. Some methods are to analyze the frequency during the driving process and judge the frequency distribution at the moment of collision to determ...

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): G06N3/08G06N3/04
CPCG06N3/08G06N3/045
Inventor 杨乐超江勇林辉潘钟声温煦
Owner 亚美智联数据科技有限公司
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
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