Unlock instant, AI-driven research and patent intelligence for your innovation.

Vehicle Classification Method Combining Gaussian Background Modeling and Recurrent Neural Network

A technology of cyclic neural network and Gaussian background, which is applied in the field of computer vision classification, can solve the problems that the Gaussian mixture model cannot overcome the false detection and the recognition accuracy needs to be improved.

Active Publication Date: 2019-08-27
NANJING UNIV
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The technical problem to be solved by the present invention is: the existing technology cannot overcome the false detection caused by the Gaussian mixture model under the change of illumination and the shaking of branches, and the recognition accuracy needs to be improved

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 Classification Method Combining Gaussian Background Modeling and Recurrent Neural Network
  • Vehicle Classification Method Combining Gaussian Background Modeling and Recurrent Neural Network
  • Vehicle Classification Method Combining Gaussian Background Modeling and Recurrent Neural Network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0031] The invention provides a vehicle classification method and system combining a Gaussian background modeling and a cyclic neural network, aiming at effectively and accurately classifying vehicle types in complex expressway scenes, and improving classification accuracy. The invention can be applied to occasions such as expressway monitoring systems and the like, and has good practicability. In the following, the present invention will be described in more detail and concretely with reference to the accompanying drawings and examples.

[0032] The first step is to model the mixed Gaussian background and extract moving objects. Such as figure 1 ,Specific steps are as follows:

[0033] 1. Initialize the highway background, first use the first n frames of continuous video stream images of the video to construct the highway background.

[0034] 2. Use K Gaussian distributions to approximate the gray value of each pixel in each frame of the image (the K value is generally 3-5...

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 video car classification method combining Gaussian background modeling and cyclic neural network uses the mixed Gaussian model to extract moving objects, and sends the moving objects to the cyclic neural network to extract features, and judges whether the target is a vehicle and the vehicle according to the vector output by the cyclic neural network. Type. The present invention proposes to use RNN as a follow-up operation of the Gaussian mixture model to achieve the purpose of vehicle classification. First, the Gaussian mixture model is used to carry out background modeling of the video sequence, and the moving target area is detected, and the detected target area is detected by CNN. Carry out classification, and input the classification results into the RNN network to obtain the final classification to determine whether it is a passenger car, a truck or a non-car. The present invention creatively uses a Gaussian background modeling method combined with a cyclic neural network. The method has strong robustness, and the combination of the two can greatly improve the accuracy of vehicle detection and vehicle type recognition.

Description

technical field [0001] The invention relates to computer vision classification technology, in particular to a method for realizing vehicle classification by using Gaussian background modeling combined with a cyclic neural network. Background technique [0002] With the rapid development of society and economy, Intelligent Transportation System (ITS) plays an increasingly important role in traffic management. Traffic parameters such as traffic volume and average speed collected by the ITS system can provide a reliable basis for the analysis and management of the traffic management department. The traditional vehicle detection method is to use induction coils to collect traffic parameters. This method is easy to damage the road surface and is troublesome to install and maintain. The vision-based video detection technology can not only collect traffic parameters, but also classify vehicle types. Moreover, the vehicle detection technology of surveillance video is one of the im...

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): G06T7/254G06K9/62G06K9/46G06K9/00
CPCG06T7/254G06T2207/20224G06T2207/10016G06V20/54G06V10/443G06F18/24G06F18/214
Inventor 阮雅端储新迪陈金艳赵博睿许山陈启美
Owner NANJING UNIV