Temporal and spatial feature based approaches in drowsiness detection using deep learning technique

Pandey, Nageshwar Nath; Muppalaneni, Naresh Babu

doi:10.1007/s11554-021-01114-x

Cited by 18 publications

(6 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The experiments involve the collection of EEG signals from eight participants in both alert and fatigue stages. The research presented by [ 55 ] focuses on two distinct categories of videos: alert and drowsy. The study utilizes a thorough dataset consisting of 60 individuals who have been classified into three groups: alert, low vigilant, and drowsy.…”

Section: Literature Reviewmentioning

confidence: 99%

Ultra-Wide Band Radar Empowered Driver Drowsiness Detection with Convolutional Spatial Feature Engineering and Artificial Intelligence

Siddiqui,

Akmal,

Iqbal

et al. 2024

Sensors

View full text Add to dashboard Cite

Driving while drowsy poses significant risks, including reduced cognitive function and the potential for accidents, which can lead to severe consequences such as trauma, economic losses, injuries, or death. The use of artificial intelligence can enable effective detection of driver drowsiness, helping to prevent accidents and enhance driver performance. This research aims to address the crucial need for real-time and accurate drowsiness detection to mitigate the impact of fatigue-related accidents. Leveraging ultra-wideband radar data collected over five minutes, the dataset was segmented into one-minute chunks and transformed into grayscale images. Spatial features are retrieved from the images using a two-dimensional Convolutional Neural Network. Following that, these features were used to train and test multiple machine learning classifiers. The ensemble classifier RF-XGB-SVM, which combines Random Forest, XGBoost, and Support Vector Machine using a hard voting criterion, performed admirably with an accuracy of 96.6%. Additionally, the proposed approach was validated with a robust k-fold score of 97% and a standard deviation of 0.018, demonstrating significant results. The dataset is augmented using Generative Adversarial Networks, resulting in improved accuracies for all models. Among them, the RF-XGB-SVM model outperformed the rest with an accuracy score of 99.58%.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Ultra-Wide Band Radar Empowered Driver Drowsiness Detection with Convolutional Spatial Feature Engineering and Artificial Intelligence

Siddiqui,

Akmal,

Iqbal

et al. 2024

Sensors

View full text Add to dashboard Cite

show abstract

“…Along the same lines, ref. [31] used temporal and spatial face features to detect drowsiness. The UTA-RLDD dataset used for experimentation has 30 h of videos of 60 participants with three classes: alert, low vigilant, and drowsy.…”

Section: Literature Reviewmentioning

confidence: 99%

“…This model begins with the initial Conv2D layer, which produces an output with dimensions (None, 62, 62, 32) and has 896 parameters. This is followed by a MaxPooling2D layer, which reduces the spatial dimensions of the feature maps to (None, 31,31,32) with no additional parameters. The subsequent layers continue this pattern of alternating Conv2D and MaxPooling2D layers, progressively reducing the spatial dimensions and increasing the number of feature maps.…”

Section: Proposed Deep Learning Architecturesmentioning

confidence: 99%

“…A deep CNN-RNN hybrid model combines the capabilities of CNN and RNN to handle sequential data and leverage spatial and temporal relationships. To capture temporal information in the data, hybrid models include recurrent layers such as LSTM and convolutional layers [31]. As a result, they are compatible with applications that need sequential data, such as video analysis and speech recognition.…”

Section: Hybrid Cnn-rnn Architecturementioning

confidence: 99%

See 1 more Smart Citation

Detection of Drowsiness among Drivers Using Novel Deep Convolutional Neural Network Model

Majeed,

Shafique,

Safran

et al. 2023

Sensors

View full text Add to dashboard Cite

Detecting drowsiness among drivers is critical for ensuring road safety and preventing accidents caused by drowsy or fatigued driving. Research on yawn detection among drivers has great significance in improving traffic safety. Although various studies have taken place where deep learning-based approaches are being proposed, there is still room for improvement to develop better and more accurate drowsiness detection systems using behavioral features such as mouth and eye movement. This study proposes a deep neural network architecture for drowsiness detection employing a convolutional neural network (CNN) for driver drowsiness detection. Experiments involve using the DLIB library to locate key facial points to calculate the mouth aspect ratio (MAR). To compensate for the small dataset, data augmentation is performed for the ‘yawning’ and ‘no_yawning’ classes. Models are trained and tested involving the original and augmented dataset to analyze the impact on model performance. Experimental results demonstrate that the proposed CNN model achieves an average accuracy of 96.69%. Performance comparison with existing state-of-the-art approaches shows better performance of the proposed model.

show abstract

“…One of the leading causes of mortality worldwide is a traffic collision, which is often brought on by the driver's exhaustion and tendency to doze off behind the wheel. Pandey and Muppalaneni (2021) used machine learning and deep learning to the two distinct models using temporal and geographical features. Long short-term memory (LSTM) and computer vision methods are used in one model to extract temporal characteristics, while convolution neural networks (CNN) and LSTM are used in the other model to extract spatial features.…”

Section: Introductionmentioning

confidence: 99%

Data-driven approach for effective fatigue detection in drivers during driving events

Vyas,

Tyagi,

Sharma

2023

Multidiscip. Sci. J.

View full text Add to dashboard Cite

The effectiveness of transportation and the safety of the road depend heavily on drivers during driving events. The way people drive, and the likelihood of accidents or other occurrences may both be strongly impacted by their actions and behaviors. Promoting safe and sensible driving behaviors requires an understanding of the traits and variables that affect drivers during driving incidents. The rising number of traffic accidents highlights the critical necessity to rein in and lessen the prevalence of careless driving. One of the most common causes of these serious mistakes is drowsiness while driving. To combat this problem, algorithms have been created to identify signs of driver weariness and sound an alarm. The developed algorithms have a serious flaw in their accuracy, and it also takes too long to identify driver drowsiness before alerting them. Timeliness and precision are two of the most important factors in preventing traffic mishaps. Multiple datasets have been utilized to improve methods for identifying signs of exhaustion or drowsiness. These data were acquired either through video streaming records of the driver's behavior or from the driver's brain electroencephalogram (EEG) readings. In order to create a high-performance fatigue detection system, this research designs a novel firefly-integrated optimum cascaded convolutional neural network (FI-OCCNN). The suggested approach offers the greatest detection accuracy among the existing classifiers, up to 98.75%. The studies further show that the recommended methods provide the greatest level of detecting precision with the quickest testing time (TT) compared to all other existing and successful tiredness detection techniques.

show abstract

Temporal and spatial feature based approaches in drowsiness detection using deep learning technique

Cited by 18 publications

References 28 publications

Ultra-Wide Band Radar Empowered Driver Drowsiness Detection with Convolutional Spatial Feature Engineering and Artificial Intelligence

Ultra-Wide Band Radar Empowered Driver Drowsiness Detection with Convolutional Spatial Feature Engineering and Artificial Intelligence

Detection of Drowsiness among Drivers Using Novel Deep Convolutional Neural Network Model

Data-driven approach for effective fatigue detection in drivers during driving events

Contact Info

Product

Resources

About