Detection of Dangerous Driver Health Problems Using HOG-Autoencoder

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27240%2F23%3A10254726" target="_blank" >RIV/61989100:27240/23:10254726 - isvavai.cz</a>

Result on the web

<a href="https://link.springer.com/chapter/10.1007/978-3-031-40971-4_43" target="_blank" >https://link.springer.com/chapter/10.1007/978-3-031-40971-4_43</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-031-40971-4_43" target="_blank" >10.1007/978-3-031-40971-4_43</a>

Result language

angličtina

Original language name

Detection of Dangerous Driver Health Problems Using HOG-Autoencoder

Original language description

In this paper, we present a method that can be used to detect unexpected driver health problems (e.g. stroke, heart attack, epileptic or similar types of seizures). Obviously, in such cases, the goal is to obtain the recognition results in the shortest possible time. Therefore, the main contribution of the presented method is the speed combined with satisfactory detection results. To achieve these goals, we use the HOG method for fast image feature extraction in the first step. In the second step, an autoencoder network is used to compress the features. Based on the autoencoder reconstruction error, it is then decided whether the driver&apos;s health condition is normal or abnormal. The results seem to be promising for the possible practical deployment.

Czech name

—

Czech description

—

Type

D - Article in proceedings

CEP classification

—

OECD FORD branch

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project

—

Continuities

S - Specificky vyzkum na vysokych skolach

Publication year

2023

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection

Lecture Notes on Data Engineering and Communications Technologies. Volume 182

ISBN

978-3-031-40970-7

ISSN

2367-4512

e-ISSN

2367-4520

Number of pages

11

Pages from-to

454-464

Publisher name

Springer

Place of publication

Cham

Event location

Čiang Mai

Event date

Sep 6, 2023

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

—