DC Field | Value | Language |
dc.contributor.author | CHACHOUA, AMani | - |
dc.contributor.author | DJABER, ROfaida | - |
dc.date.accessioned | 2024-10-03T09:26:10Z | - |
dc.date.available | 2024-10-03T09:26:10Z | - |
dc.date.issued | 2024 | - |
dc.identifier.uri | https://repository.esi-sba.dz/jspui/handle/123456789/705 | - |
dc.description | Supervisor : Pr. Sidi Mohammed BENSLIMANE Co-supervisor : Dr. Rabab BOUSMAHA | en_US |
dc.description.abstract | Road traffic accidents result in significant losses of life and property, often due to factors
such as driver fatigue and drowsiness. Therefore, real-time monitoring of a driver’s state
within the vehicle and accurate detection of fatigue are crucial to reducing the number of
accidents. In recent years, ongoing advancements in computing technology and artificial
intelligence have significantly enhanced driver monitoring systems. Numerous experimental
studies have gathered real driver drowsiness data and utilized various AI algorithms
and feature combinations to improve the real-time performance of these systems. This
thesis reviews current research on driver drowsiness detection systems developed in recent
years, showcasing recent systems and approaches that use different types of measures to
detect drowsiness and categorizing each system based on the type of information used.
Each system discussed in this thesis includes a detailed description of the features,
classification algorithms, and datasets utilized. Additionally, the thesis evaluates these
systems in terms of accuracy, sensitivity, and precision. Furthermore, the thesis illustrates
the challenges in the field of driver drowsiness detection and presents some future trends
in the field. ****
Les accidents de la route entraˆınent d’importantes pertes en vies humaines et en biens,
souvent dues `a des facteurs tels que la fatigue et la somnolence du conducteur. Par
cons´equent, la surveillance en temps r´eel de l’´etat du conducteur dans le v´ehicule et la
d´etection pr´ecise de la fatigue sont essentielles pour r´eduire le nombre d’accidents. Ces
derni`eres ann´ees, les progr`es continus de la technologie informatique et de l’intelligence
artificielle ont consid´erablement am´elior´e les syst`emes de surveillance des conducteurs. De
nombreuses ´etudes exp´erimentales ont rassembl´e des donn´ees r´eelles sur la somnolence
des conducteurs et utilis´e divers algorithmes d’IA et combinaisons de fonctionnalit´es pour
am´eliorer les performances en temps r´eel de ces syst`emes. Cette th`ese passe en revue
les recherches actuelles sur les syst`emes de d´etection de la somnolence des conducteurs
d´evelopp´es ces derni`eres ann´ees, en pr´esentant des syst`emes et des approches r´ecents qui
utilisent diff´erents types de mesures pour d´etecter la somnolence et en cat´egorisant chaque
syst`eme en fonction du type d’informations utilis´ees.
Chaque syst`eme discut´e dans cette th`ese comprend une description d´etaill´ee des
fonctionnalit´es, des algorithmes de classification et des ensembles de donn´ees utilis´es. De
plus, la th`ese ´evalue ces syst`emes en termes d’exactitude, de sensibilit´e et de pr´ecision.
De plus, la th`ese illustre les d´efis dans le domaine de la d´etection de la somnolence des
conducteurs et pr´esente quelques tendances futures dans le domaine. | en_US |
dc.language.iso | en | en_US |
dc.subject | Driver Drowsiness Detection | en_US |
dc.subject | Artificial Intelligence | en_US |
dc.subject | Machine Learning | en_US |
dc.subject | Deep Learning | en_US |
dc.subject | Computer Vision | en_US |
dc.subject | Transfer Learning | en_US |
dc.subject | Biological-based Measures | en_US |
dc.subject | Hybrid-based Measures | en_US |
dc.subject | Vehicle-based Measures | en_US |
dc.title | Comparative Analysis Of State-Of-The-Art Techniques In Driver Drowsiness Detection System | en_US |
dc.type | Thesis | en_US |
Appears in Collections: | Master
|