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

A Vehicle Detection Method in Video Stream

A technology of vehicle detection and video streaming, which is applied in the field of vehicle detection in video streaming, can solve problems such as inability to detect new vehicles, tracking loss, etc., and achieve the effects of tracking speed and stability, reducing missed detection rate, and high precision

Active Publication Date: 2021-11-30
DALIAN UNIV OF TECH
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The disadvantage of this is that it is impossible to detect new vehicles, and if the ambient light of the vehicle changes greatly, such as when entering and exiting a tunnel, it is easy to lose tracking

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
  • A Vehicle Detection Method in Video Stream
  • A Vehicle Detection Method in Video Stream
  • A Vehicle Detection Method in Video Stream

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0058] The invention is further described below with reference to the accompanying drawings.

[0059] figure 1 A total algorithm flow chart is given. For the first frame of the video stream, the position of the vehicle on the image is locked using the vehicle detection algorithm, and each vehicle is assigned a filter for tracking. Considering that a new vehicle may appear in the video stream, vehicle detection, refresh filter number is performed and the filter group parameters are performed for each frame. Finally, the association of the data association policy guarantees the association of the interframe vehicle.

[0060] The convolutional neural network structure and parameters used in the vehicle detection portion in step A3 are shown in the following table:

[0061]

[0062] figure 2 The algorithm flow diagram of how to filter success match data pairs is given. The input data includes two parts, detection results, and tracking results. Where O 1 O 2 , ..., o n Indicates 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 invention discloses a vehicle detection method in a video stream, comprising the following steps: acquiring a video frame, using a trained convolutional neural network to identify the obtained video frame, and judging whether there is a vehicle in the video according to the result, and the vehicle The specific position of the vehicle; the position of the vehicle in the image is obtained by the vehicle detection, and its value in the next frame is predicted by the Kalman filter as the vehicle tracking result; the results of the vehicle detection and the results of the vehicle tracking are correlated, according to the correlation results Modify filter parameters and update filter group information. The invention designs a convolutional neural network with less network layers and a small structure for the vehicle detection task, which has fast detection speed and high precision. The invention uses a Kalman tracking algorithm to predict the position of the detection frame, and the tracking speed is fast and stable. The invention uses a data association strategy to combine a detection algorithm with a tracking algorithm, embodies the correlation of vehicles between frames, and can reduce the missed detection rate of vehicle objects.

Description

Technical field [0001] The present invention relates to a vehicle detecting algorithm based on computer vision, particularly a vehicle detection method in a video stream. Background technique [0002] Objection detection refers to the use of a computer visual algorithm to find all of the exclusive vehicles in the image, determine their position and category. Widely used in video surveillance, no driving, human-machine interaction and other fields. [0003] The vehicle detection method in the currently known video stream is mainly divided into two categories. One is to divide the video stream into each frame image, and then detect the vehicle in each frame image. The drawbacks of this are that the amount of calculation is large, the efficiency is low, and does not reflect the association of the vehicle between the frame and the frame, resulting in a lower level of intelligence. [0004] The other category is to detect the vehicle in the first frame, and then several frames are tra...

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): G06K9/00G06K9/46G06K9/62G06N3/04
Inventor 郭烈何丹妮郑仁成姚宝珍李琳辉孙大川
Owner DALIAN UNIV OF TECH
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