Smart Parking

Peter Stanchev; John Geske

doi:10.2205/CODATA2017

Smart Parking

Peter Stanchev, John Geske

Kettering University

Research output: Contribution to journal › Article › peer-review

Abstract

The paper presents an efficient solution for real-time parking lot occupancy detection based on Convolutional Neural Network classifier, real time image segmentation and analysis, and streaming data. It takes in account different light conditions, parts of the day, and seasons. It has been used benchmarks collections for parking occupancy detection. Problems that we solved are: significant changes of lighting conditions - sunny, rainy and snowing days; different time of the day; partially occupant, moving cars and peoples, additional objects.

Original language	American English
Journal	Geoinformatics Research Papers
Volume	5
DOIs	https://doi.org/10.2205/CODATA2017
State	Published - 2017

Keywords

Neural Network classifier
image segmentation
Smart Parking

Disciplines

Computer Sciences

Access to Document

10.2205/CODATA2017License: Unspecified

CODATA2017.pdfFinal published version, 42.6 KBLicense: Unspecified

Cite this

@article{fc7bacc492454d24835363248f459863,

title = "Smart Parking",

abstract = " The paper presents an efficient solution for real-time parking lot occupancy detection based on Convolutional Neural Network classifier, real time image segmentation and analysis, and streaming data. It takes in account different light conditions, parts of the day, and seasons. It has been used benchmarks collections for parking occupancy detection. Problems that we solved are: significant changes of lighting conditions - sunny, rainy and snowing days; different time of the day; partially occupant, moving cars and peoples, additional objects. ",

keywords = "Neural Network classifier, image segmentation, Smart Parking",

author = "Peter Stanchev and John Geske",

year = "2017",

doi = "10.2205/CODATA2017",

language = "American English",

volume = "5",

journal = "Geoinformatics Research Papers",

}

TY - JOUR

T1 - Smart Parking

AU - Stanchev, Peter

AU - Geske, John

PY - 2017

Y1 - 2017

N2 - The paper presents an efficient solution for real-time parking lot occupancy detection based on Convolutional Neural Network classifier, real time image segmentation and analysis, and streaming data. It takes in account different light conditions, parts of the day, and seasons. It has been used benchmarks collections for parking occupancy detection. Problems that we solved are: significant changes of lighting conditions - sunny, rainy and snowing days; different time of the day; partially occupant, moving cars and peoples, additional objects.

AB - The paper presents an efficient solution for real-time parking lot occupancy detection based on Convolutional Neural Network classifier, real time image segmentation and analysis, and streaming data. It takes in account different light conditions, parts of the day, and seasons. It has been used benchmarks collections for parking occupancy detection. Problems that we solved are: significant changes of lighting conditions - sunny, rainy and snowing days; different time of the day; partially occupant, moving cars and peoples, additional objects.

KW - Neural Network classifier

KW - image segmentation

KW - Smart Parking

U2 - 10.2205/CODATA2017

DO - 10.2205/CODATA2017

M3 - Article

VL - 5

JO - Geoinformatics Research Papers

JF - Geoinformatics Research Papers

ER -