| Autor: |
null Kenechukwu Sylvanus Anigbogu, null Hyacinth Chibueze Inyiama, null Ikechukwu Onyenwe, null Sylvanus Okwudili Anigbogu |
| Rok vydání: |
2022 |
| Předmět: |
|
| Zdroj: |
World Journal of Advanced Engineering Technology and Sciences. 7:137-148 |
| ISSN: |
2582-8266 |
| DOI: |
10.30574/wjaets.2022.7.1.0103 |
| Popis: |
Driving is a complex and dynamic task requiring drivers not only to make accurate perceptions and cognitions about the information on the driver’s driving skill but also to process this information at a high speed. This paper compared three major image processing/machine learning algorithms viz; Single Shot Multibox Detection (SSD), Convolutional Neural Networks (CNN), and support vector machine (SVM) to find the fastest and most efficient of the three with regards to the dataset from driving events (braking, speeding and safe driving) collected from Nigeria. The results analyzed showed that in an identical testing environment, Support Vector Machine outperformed Single Shot Detection and Convolutional Neural Networks. |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
