Passenger number counting method based on behavior characteristics

Abstract

Disclosed is a passenger number counting method based on behavior characteristics. The behaviors when people go on and off a car are divided into four behavior modes, then an analysis study mode based on SVM is adopted, the kernel function of the four behavior modes is firstly determined, training data are mapped to high-dimensional space through a mapping phi, and an optimal hyperplane is solved so that the training data can be linearly detachable. The feature points of the four behavior modes are extracted, corresponding feature matrixes are generated, and a classifier is built. When a group of unknown data needs to be analyzed, the data are input to the classifier to be classified, the final classification result is the behavior mode corresponding to the group of data, and passengers can be counted according to the behavior modes. The passenger number counting method based on behavior characteristics integrates the behavior features and the computer technology according to the characteristics of an intelligent traffic identification technology, improves the measurement accuracy and real-time performance, obtains more actual, accurate and real-time loading information of the car, and provides important reference for management and monitoring of the car.

Description

technical field [0001] The invention belongs to the technical field of intelligent vehicle control, and relates to a method for measuring the number of passengers based on behavioral characteristics. The method judges the behavior state of people to count the number of people getting on and off the bus and the current number of passengers. Background technique [0002] With the development of the economy, people's travel methods have become diversified, and traveling by car is still one of the main travel methods for people. Travel by car can be roughly divided into the following three categories: travel by private car, travel by bus, and travel by coach. [0003] As our country vigorously promotes green travel, more and more people choose the latter two travel methods, but some problems that follow cannot be ignored. One of the very serious problems is traffic congestion. One of the main reasons is that the uneven development and inconsistent growth rate of public transpo...

AI Technical Summary

Problems solved by technology

However, the bills cannot directly reflect the number of passengers. Some vehicles even adopt the method of getting on the bus first and paying the fare later.

In terms of passenger flow statistics, most of the existing patented technologies use technologies such as RFID or image recognition methods. The measurement form is single and does not reflect better intelligence. Moreover, the measured real-time vehicle information is not accurate, and relevant departments cannot obtain reliable data.

Images