Рақамли технологияларнинг назарий ва амалий масалалари Volume 7 Issue 3 (2024) · pp. 131-136
Intelligent algorithms for determining physiological states of yawning for fatigue assessment
Назаров, Файзулло, Хамидов, Мунис
Abstract
In this study, yawning, which is considered one of the physiological deviations of humans, mouth opening state classification and several machine learning algorithms were used. The CNN model outperformed traditional machine learning methods such as SVM, MLP and KNN with an accuracy of 97.07%. In the study, a dataset was formed using mouth part images extracted from face images using CNN and the model was trained. In addition, the overall performance of the model is improved by dataset normalization and dropout methods. The results demonstrate the effectiveness of the CNN algorithm in detecting the mouth opening state (yawning) and open the possibility for further improvement of fatigue state detection systems in the future.
обнаружение зеваниясверточные нейронные сети (CNN)машинное обучениеклассификация изображенийизвлечение признаков изображенияyawn detectionconvolutional neural networks (CNN)machine learningimage classificationimage feature extraction
Metadata source: the journal's OAI-PMH archive · Sindex does not store the full text; it links to the source.