Modern drowsiness detection techniques: a review

Jasim, Sarah Saadoon; Hassan, Alia Karim Abdul

doi:10.11591/ijece.v12i3.pp2986-2995

Cited by 8 publications

(12 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Statistical imputation techniques such as mean and mode imputation for categorical features to maintain the integrity and efficacy of the detection model features with a high percentage of missing values can also be excluded using feature selection approaches. [16][17][18]…”

Section: Data Cleaning and Pre-processingmentioning

confidence: 99%

“…The authors also suggest looking into how other hyperparameters, such as batch size and learning rate, affect the effectiveness of the DW-FedAvg approach. Further investigation may examine the DW-FedAvg approach's application to domains [18] other than Android malware classification to assess its efficacy in various contexts [46].…”

Section: Behavioural Analysis Modelmentioning

confidence: 99%

See 1 more Smart Citation

Mobile based Malware Detection using Artificial Intelligence Techniques a review

Faez mahdi,

Saadoon Jasim

2024

J. Al-Qadisiyah Comp. Sci. Math.

View full text Add to dashboard Cite

Malware attacks on mobile devices are becoming more common and more complicated every year. Malware writers see the open-source Android app as their main target because so many people use it. Artificial intelligence is used by most of the literature's mobile malware detection methods to find ransomware. Our research, on the other hand, makes it clear that most of the earlier studies used different metrics and models, as well as different datasets and classification features that came from static, dynamic, or hybrid analysis strategies. This makes comparing the different suggested detection methods more difficult and may also make the results less certain. The goal of this work is to solve the problem of AI-powered malware detection by sorting current methods and approaches into three groups: the type of dataset, the type of detection method used (machine learning models, deep learning models, and Behavioral Analysis model), and how well the method works. In this way, we suggest a convergent plan that can be used as a basis for future methods of finding malware on Android and as a solid standard for artificial intelligence work in this area.

show abstract

Section: Data Cleaning and Pre-processingmentioning

confidence: 99%

Section: Behavioural Analysis Modelmentioning

confidence: 99%

Mobile based Malware Detection using Artificial Intelligence Techniques a review

Faez mahdi,

Saadoon Jasim

2024

J. Al-Qadisiyah Comp. Sci. Math.

View full text Add to dashboard Cite

show abstract

“…Several survey and review studies that integrated DL models with IoT devices in driver drowsiness detection systems [ 5 , 6 , 7 , 8 ] have reported that the major challenge is training the DL models, in that DL has a complex architecture and is heavyweight. The approximate range of the DL model size in the literature on driver drowsiness detection is from 10 MB to 54 MB.…”

Section: Introductionmentioning

confidence: 99%

DDD TinyML: A TinyML-Based Driver Drowsiness Detection Model Using Deep Learning

Alajlan

Ibrahim

2023

Sensors

View full text Add to dashboard Cite

Driver drowsiness is one of the main causes of traffic accidents today. In recent years, driver drowsiness detection has suffered from issues integrating deep learning (DL) with Internet-of-things (IoT) devices due to the limited resources of IoT devices, which pose a challenge to fulfilling DL models that demand large storage and computation. Thus, there are challenges to meeting the requirements of real-time driver drowsiness detection applications that need short latency and lightweight computation. To this end, we applied Tiny Machine Learning (TinyML) to a driver drowsiness detection case study. In this paper, we first present an overview of TinyML. After conducting some preliminary experiments, we proposed five lightweight DL models that can be deployed on a microcontroller. We applied three DL models: SqueezeNet, AlexNet, and CNN. In addition, we adopted two pretrained models (MobileNet-V2 and MobileNet-V3) to find the best model in terms of size and accuracy results. After that, we applied the optimization methods to DL models using quantization. Three quantization methods were applied: quantization-aware training (QAT), full-integer quantization (FIQ), and dynamic range quantization (DRQ). The obtained results in terms of the model size show that the CNN model achieved the smallest size of 0.05 MB using the DRQ method, followed by SqueezeNet, AlexNet MobileNet-V3, and MobileNet-V2, with 0.141 MB, 0.58 MB, 1.16 MB, and 1.55 MB, respectively. The result after applying the optimization method was 0.9964 accuracy using DRQ in the MobileNet-V2 model, which outperformed the other models, followed by the SqueezeNet and AlexNet models, with 0.9951 and 0.9924 accuracies, respectively, using DRQ.

show abstract

“…And biosensors (such as electromyography (EMG) [5], as well as force sensors [6], Electroencephalography (EEG), and magnetoencephalography (MEG) have been used. Electromechanical sensors can effectively detect biological data [5], [7].…”

Section: Introductionmentioning

confidence: 99%

A survey on bio-signal analysis for human-robot interaction

Radha

Hassan

2022

IJECE

View full text Add to dashboard Cite

<span lang="EN-US">The use of bio-signals analysis in human-robot interaction is rapidly increasing. There is an urgent demand for it in various applications, including health care, rehabilitation, research, technology, and manufacturing. Despite several state-of-the-art bio-signals analyses in human-robot interaction (HRI) research, it is unclear which one is the best. In this paper, the following topics will be discussed: robotic systems should be given priority in the rehabilitation and aid of amputees and disabled people; second, domains of feature extraction approaches now in use, which are divided into three main sections (time, frequency, and time-frequency). The various domains will be discussed, then a discussion of each domain's benefits and drawbacks, and finally, a recommendation for a new strategy for robotic systems.</span>

show abstract

Modern drowsiness detection techniques: a review

Cited by 8 publications

References 41 publications

Mobile based Malware Detection using Artificial Intelligence Techniques a review

Mobile based Malware Detection using Artificial Intelligence Techniques a review

DDD TinyML: A TinyML-Based Driver Drowsiness Detection Model Using Deep Learning

A survey on bio-signal analysis for human-robot interaction

Contact Info

Product

Resources

About