Work

Airport Surface Targets Detection Based on YOLO v3

Year: 2023

Type: conference-paper

Abstract: With the rapid development of civil aviation in recent years, the number of flights has increased, the airport is busy, small targets and aircraft are likely to block each other are difficult to detect, and there is a risk of conflict. The real-time YOLO algorithm is applied to airport scene aircraft detection, and the convolution layer is added to the network to improve the detection effect by adapting the detector to the projection image of the monitoring system. The results show that the target frame rate of airport detection based on YOLO optimization is 1.1 frames / s, the accuracy rate is over 80%, and it is robust to the environment, so as to provide the decision basis for operation and dispatch center.

Source: International Conference on Machine Learning and Computing (ICMLC)

DOI: 10.1145/3587716.3587766

Authors: Jun Li, Weizhen Tang, Mingming Yang

Institutions: Civil Aviation Flight University of China, Chinese Academy of Civil Aviation Science and Technology

Language: English

Topic: Advanced Neural Network Applications

Subfield: Computer Vision and Pattern Recognition

Field: Computer Science

Domain: Physical Sciences

Open Access status: bronze

OpenAlex ID: W4386502153

FWCI: 0.107

Cites: 3

Cited by: 1

Related to: 10