undefined

A Paradigm for Object Detection to Recognize and Classify Vehicles Using Computer Vision

Publiceringsår

2025

Upphovspersoner

Pitkäkangas, Ville; Kaakinen, Heikki; Tuunainen, Tom; Isohanni, Olli; Jose, Mitha Rachel

Abstrakt

Computer vision has emerged as a game-changing technology in the mining industry, revolutionizing operations and unlocking various use case scenarios. With increased trade facilities, ports are recognized as one of the most diligent work environments globally. The applications of machine learning and computer vision in ports offer improved security, efficient container management, intelligent traffic management, predictive maintenance, automated operations, and environmental monitoring. These advancements contribute to streamlined processes, cost reduction, enhanced safety, and overall optimization in port environments. This study proposes an approach to detect and classify vehicles in a port during the wintertime in Finland using computer vision and machine learning methods. Due to the high variability between seasons, particularly winter and summer in Finland, there might be a need to categorize images by time of year. The study is developed as a model to detect and classify vehicles in the port area, and the port used in the study acts as an international hub for various trades and industries, including but not limited to chemistry and mining.
Visa mer

Organisationer och upphovspersoner

Centria-ammattikorkeakoulu

Kaakinen Heikki Orcid -palvelun logo

Isohanni Olli

Tuunainen Tom Orcid -palvelun logo

Pitkäkangas Ville Orcid -palvelun logo

Publikationstyp

Publikationsform

Artikel

Moderpublikationens typ

Tidning

Artikelstyp

En originalartikel

Målgrupp

Vetenskaplig

Kollegialt utvärderad

Kollegialt utvärderad

UKM:s publikationstyp

A1 Originalartikel i en vetenskaplig tidskrift

Publikationskanalens uppgifter

Öppen tillgång

Öppen tillgänglighet i förläggarens tjänst

Ja

Öppen tillgång till publikationskanalen

Helt öppen publikationskanal

Licens för förläggarens version

CC BY NC ND

Parallellsparad

Nej

Övriga uppgifter

Vetenskapsområden

Data- och informationsvetenskap

Nyckelord

[object Object],[object Object],[object Object],[object Object]

Publiceringsland

Jordanien

Förlagets internationalitet

Internationell

Språk

engelska

Internationell sampublikation

Nej

Sampublikation med ett företag

Nej

Publikationen ingår i undervisnings- och kulturministeriets datainsamling

Ja