Work

A Spatiotemporal Feature-Based Approach for the Detection of Unlicensed Taxis in Urban Areas

Year: 2024

Type: article

Abstract: Unlicensed taxis seriously disrupt the transportation market order, and threaten passenger safety. Therefore, this paper proposes a method for identifying unlicensed taxis based on travel characteristics. First, the vehicle mileage and operation time are calculated using traffic surveillance bayonet data, and variance analysis is applied to identification indicators for unlicensed taxis. Secondly, the mathematical model for identifying unlicensed taxis is established. The model is validated using the Hosmer-Lemeshow test, confusion matrix and ROC curve analysis. Finally, by applying methods such as geographic information matching, the spatiotemporal distribution characteristics of suspected unlicensed taxis in a city in Anhui Province are identified. The results show that the model effectively identifies suspected unlicensed taxis (ACC = 99.10%). The daily average mileage, daily average operating time, and number of operating days for suspected unlicensed taxis are significantly higher than those for private cars. Additionally, the suspected unlicensed taxis exhibit regular patterns in their travel origin-destination points and temporal distribution, enabling traffic management authorities to implement targeted regulatory measures.

Source: Sensors

DOI: 10.3390/s24248206

Authors: Yun Xiao, Rongqiao Li, Jinyan Li

Institutions: Hefei University, Northeast Forestry University

Language: English

Topic: Traffic Prediction and Management Techniques

Subfield: Building and Construction

Field: Engineering

Domain: Physical Sciences

SDGs: Sustainable cities and communities, Industry, innovation and infrastructure, Peace, justice, and strong institutions

Open Access status: gold

Estimated APC paid: $2,955