Expressway traffic incident detection method based on optimized support vector machine (SVM)

Abstract

The invention discloses an expressway traffic incident detection method based on an optimized support vector machine (SVM). The expressway traffic incident detection method comprises the following steps of: quickly and accurately acquiring traffic flow parameters through an expressway camera video, preprocessing the data, and classifying the data into a training data set and a test data set; selecting a radial basis function (RBF) function by a SVM model, and performing optimization selection on a punishment parameter C and a core parameter gamma of the SVM model by adopting an improved network search algorithm; training the SVM model through the training data set; and checking the performance of a trained SVM incident detection model through the test data set, and detecting a real-time traffic incident on an expressway by the SVM model. The test shows that the detection rate is over 90 percent and the error alarm rate is below 5 percent; and the optimization time is short, so that the requirement on the instantaneity of detecting the traffic incident is met.

Description

Technical field [0001] The invention belongs to the field of machine learning and data mining, and is mainly used in a highway traffic management system, and is a highway traffic incident detection method based on optimized SVM. Background technique [0002] All along, highway traffic congestion has greatly increased people’s travel time and fuel consumption [1-2] , Brought huge losses to people. However, most traffic jams are caused by non-recurring traffic incidents [3] . Therefore, how to detect traffic incidents quickly and accurately is of great significance for reducing the impact and duration of traffic incidents and implementing reasonable road network optimization. [0003] The performance of the traffic incident detection system mainly depends on two aspects: data collection and data processing. Data collection refers to the use of some detection techniques to obtain traffic flow parameters. Data processing refers to analyzing the acquired traffic flow parameters throu...

AI Technical Summary

Problems solved by technology

However, coil detectors also have some disadvantages: coil detectors need to be installed under each lane, which is expensive and cannot cover a large area; when traffic is congested and the distance between vehicles is less than 3m, the accuracy of the detector is greatly reduced, and even Unable to detect; the slit of the embedded coil softens the road surface and makes the road surface vulnerable. When there is a problem with the road surface, it is easy to cause damage to the coil. During maintenance, it is necessary to close the driveway and excavate the road surface. Ability has some influence [7]

The event detection algorithm based on the artificial neural network is an intelligent traffic event detection algorithm, which has the advantages of high detection rate and low false alarm rate, but there is no unified criterion for the determination of the neural network structure, which requires a large number of learning samples and is prone to Over-fitting phenomenon, the generalization ability of the model is not good; while the support vector machine has a complete statistical learning theory and excellent learning performance, it can achieve a high detection rate without a large number of learning samples, and has good generalization It is a widely used traffic incident detection method at present. [13]

Images