Vehicle detection based on raspberry pi and YOLO algorithm
XUE Peiyou
LIU Lei
LIU Ruiyang
BIAN Wenyang
YAN Dongmei
Abstract：As the number of civilian vehicles in China shows a remarkable growth trend,traffic manage-ment is facing unprecedented challenges.Against this backdrop,the development of intelligent transporta-tion systems(ITS)is extremely urgent.As a core and crucial technology of ITS,vehicle type detection plays a pivotal role in the field of traffic management.However,the existing vehicle type detection systems have obvious drawbacks such as high resource consumption and exorbitant costs,which greatly limit their wide-spread promotion and application.In response to this situation,this paper carefully constructs the TenCar dataset containing 10 common types of vehicles and proposes a low-cost vehicle type detection sys-tem based on the Raspberry Pi and the YOLOv4-tiny algorithm.Through practical verification,the YOLOv4-tiny model has achieved an accuracy rate of 85.65％and a recall rate of 99.9％on the TenCar dataset.This system skillfully utilizes image recognition technology and streamlined hardware facilities to efficiently and automatically detect the vehicle types when vehicles enter or exit a venue.This feature greatly simplifies the process of vehicle search,making management work more convenient and efficient.Moreover,the system has good popularization potential and is easy to be promoted and applied in places such as companies and residential communities.
Keywords：raspberry pivehicle detectionYOLOv4-tinyYOLOv5s
Publication Date：2025-12-31
Online Publishing Date：2026-08-28(First online date of this platform, not the publication date of the document)
Pages：13( 840-852 )