As the number of road accidents increases, it is critical to avoid making driving mistakes. Driver fatigue detection is a concern that has prompted researchers to develop numerous algorithms to address this issue. The challenge is to identify the sleepy drivers with accurate and speedy alerts. Several datasets were used to develop fatigue detection algorithms such as electroencephalogram (EEG), electrooculogram (EOG), electrocardiogram (ECG), and electromyogram (EMG) recordings of the driver’s activities e.g., DROZY dataset. This study proposes a fatigue detection system based on Fast Fourier Transform (FFT) and Discrete Wavelet Transform (DWT) with machine learning and deep learning classifiers. The FFT and DWT are used for feature extraction and noise removal tasks. In addition, the classification task is carried out on the combined EEG, EOG, ECG, and EMG signals using machine learning and deep learning algorithms including 1D Convolutional Neural Networks (1D CNNs), Concatenated CNNs (C-CNNs), Support Vector Machine (SVM), Random Forest (RF), Decision Tree (DT), k-Nearest Neighbor (KNN), Quadrature Data Analysis (QDA), Multi-layer Perceptron (MLP), and Logistic Regression (LR). The proposed methods are validated on two scenarios, multi-class and binary-class classification. The simulation results reveal that the proposed models achieved a high performance for fatigue detection from medical signals, with a detection accuracy of 90% and 96% for multiclass and binary-class scenarios, respectively. The works in the literature achieved a maximum accuracy of 95%. Therefore, the proposed methods outperform similar efforts in terms of detection accuracy.

/pdf/wft-fati-dec-enhanced-fatigue-detection-ai-system-based-on-38msqj3f.pdf

WFT-Fati-Dec: Enhanced Fatigue Detection AI System Based on Wavelet Denoising and Fourier Transform

Today, biometrics are the preferred technologies for person identification, authentication, and verification cutting across different applications and industries. Sadly, this ubiquity has invigorated criminal efforts aimed at violating the integrity of these modalities. Our study presents a multi-biometric cancellable scheme (MBCS) that exploits the proven utility of deep learning models to fuse multi-exposure fingerprint, finger vein, and iris biometrics by using an Inspection V3 pre-trained model to generate an aggregate tamper-proof cancellable template. To validate our MBCS, we employed an extensive evaluation including visual, quantitative, and qualitative assessments as well as complexity analysis where average outcomes of 99.158%, 24.523 dB, 0.079, 0.909, 59.582 and 23.627 were recorded for NPCR, PSNR, SSIM, UIQ, SD and UACI respectively. These quantitative outcomes indicate that the proposed scheme compares favourably against state-of-the-art methods reported in the literature. To further improve the utility of the proposed MBCS, we are exploring its refinement to facilitate generation of cancellable templates for real-time biometric applications in person authentication at airports, banks, etc.

/pdf/an-efficient-multi-biometric-cancellable-biometric-scheme-143mmq6tk0.pdf

An efficient multi-biometric cancellable biometric scheme based on deep fusion and deep dream

An array of features and methods are being developed over the past six decades for Speaker Identification (SI) and Speaker Verification (SV), jointly known as Speaker Recognition(SR). Mel Frequency Cepstral Coefficients (MFCC) is generally used as feature vectors in most of the cases because it gives higher accuracy compared to other features. The presented paper focuses on comparative study of state-of-the-art SR techniques along with their design challenges, robustness issues and performance evaluation methods. Rigorous experiments have been performed using Gaussian Mixture Model (GMM) with variations like Universal Background Model (UBM) and/or Vector Quantization (VQ) and/or VQ based UBM-GMM (VQ-UBM-GMM) with detail discussion. Other popular methods have been included, namely, Linear Discriminate Analysis (LDA), Probabilistic LDA (PLDA), Gaussian PLDA (GPLDA), Multi-condition GPLDA (MGPLDA), Identity Vector (i-vector) for comparative study only. Three popular audio data-sets have been used in the experiments, namely, IITG-MV SR, Hyke-2011 and ELSDSR. Hyke-2011 and ELSDSR contain clean speech while IITG-MV SR contains noisy audio data with variations in channel (device), environment, spoken style. We propose a new data mixing approach for SR to make the system independent of recording device, spoken style and environment. The accuracy we obtained for VQ and GMM based methods for databases, Hyke-2011 and ELSDSR are varies from $$99.6\%$$
 to $$100\%$$
 whereas accuracy for IITG-MV SR is upto $$98\%$$
 . Indeed, in some cases the accuracies degrade drastically due to mismatch between training and testing data as well as singularity problem of GMM. The experimental results serve as a benchmark for VQ/GMM/UBM based methods for the IITG-MV SR database.

Closed-set speaker identification using VQ and GMM based models

This research paper proposes a novel voiceprint generation methodology for recognizing the speakers registered in a system. The proposed methodology is a keyword-dependent closed set speaker classification task. The features used are Mel-Spectrogram, Chromagram, MFCC and a new ensembled feature called Mel-Chroma. Mel-Chroma is generated with the combination of Mel-spectrogram and Chromagram. The Mel-Chroma spectrogram generated is converted into a binary image by using the average as the threshold. The recurrent neural network model LSTM is used for the classification task and the dataset used is FSDD. The proposed method has a higher accuracy than the state-of-art methods for the specific task. The accuracy obtained for the classification of speakers using a binary Mel-Chroma voiceprint is 98.33%.

novel voiceprint using ensembled Mel-Chromagram for speaker recognition

Due to the widespread issue of road accidents, researchers have been drawn to investigate strategies to prevent them. One major contributing factor to these accidents is driver fatigue resulting from exhaustion. Various approaches have been explored to address this issue, with machine and deep learning proving to be effective in processing images and videos to detect asymmetric signs of fatigue, such as yawning, facial characteristics, and eye closure. This study proposes a multistage system utilizing machine and deep learning techniques. The first stage is designed to detect asymmetric states, including tiredness and non-vigilance as well as yawning. The second stage is focused on detecting eye closure. The machine learning approach employs several algorithms, including Support Vector Machine (SVM), k-Nearest Neighbor (KNN), Multi-layer Perceptron (MLP), Decision Tree (DT), Logistic Regression (LR), and Random Forest (RF). Meanwhile, the deep learning approach utilizes 2D and 3D Convolutional Neural Networks (CNNs). The architectures of proposed deep learning models are designed after several trials, and their parameters have been selected to achieve optimal performance. The effectiveness of the proposed methods is evaluated using video and image datasets, where the video dataset is classified into three states: alert, tired, and non-vigilant, while the image dataset is classified based on four facial symptoms, including open or closed eyes and yawning. A more robust system is achieved by combining the image and video datasets, resulting in multiple classes for detection. Simulation results demonstrate that the 3D CNN proposed in this study outperforms the other methods, with detection accuracies of 99 percent, 99 percent, and 98 percent for the image, video, and mixed datasets, respectively. Notably, this achievement surpasses the highest accuracy of 97 percent found in the literature, suggesting that the proposed methods for detecting drowsiness are indeed effective solutions.

/pdf/an-adaptive-fatigue-detection-system-based-on-3d-cnns-and-2oadzwn5.pdf

An Adaptive Fatigue Detection System Based on 3D CNNs and Ensemble Models

Speaker recognition is one of several biometric recognition systems owing to its high importance in numerous applications of security and telecommunications. The key aspiration of speaker recognition systems is to know who is speaking depending on voice characteristics. This paper presents an extensive study of speaker recognition in both text-dependent and text-independent cases. Convolutional Neural Network (CNN) based feature extraction is extended to the text-dependent and text-independent speaker recognition tasks. In addition, the effect of reverberation on the speaker recognition system is addressed. All speech signals are converted into images by obtaining their spectrograms. Two proposed CNN models are presented for efficient speaker recognition from clean and reverberant speech signals. They depend on image processing concepts applied on spectrograms of speech signals. One of the proposed models is compared with a conventional Benchmark model in the text-independent scenario. The performance of the recognition system is measured by the recognition rate in the cases of clean and reverberant speech.

Text-dependent and text-independent speaker recognition of reverberant speech based on CNN

Speaker recognition is one of many biometric authentications, due to its high importance in many applications of security considerations and telecommunications. The main aspiration of speaker recognition system is to know who is speaking depending on voice characteristics. Many current researches focuses on text-dependent speaker recognition which has a pre-knowledge of what utterance the speaker will say. In this paper text-independent speaker recognition system is used, where no prior knowledge is accessible in the context of the speakers’ utterances for all stages. A Convolutional Neural Network (CNN) based feature extraction is extended to a text-independent Speaker recognition task. Also the effect of reverberation on speaker recognition is addressed. All the speech signals are converted into images by obtaining their spectrograms. A proposed CNN model is presented to enhance the performance of the system in case of a reverberant signal. It depends on image processing concepts, and hence spectrograms of signals are used. The proposed model is compared with a conventional benchmark model. The performance of the recognition system is measured by the recognition rate in the case of clean and reverberant data.

Effect of Reverberation Phenomena on Text- independent Speaker Recognition Based Deep Learning

Reverberation is one of the effects that occur regularly in closed room due to multiple reflections. This paper investigates the result of reverberation on both male and female speech signals. This effect is reflected in pitch frequency of speech signals. This parameter is important as it is usually used for speaker identification. Hence, several methods for pitch frequency estimation are investigated and compared on clear and reverberant male and female speech signals to select the one that is not affected so much by the reverberation effect.

https://mjeer.journals.ekb.eg/article_69195_2bb6002764801de5d1b34d3ce2e12df5.pdf

Shaimaa E. A. Aziz Hassan

Papers

Text-dependent and text-independent speaker recognition of reverberant speech based on CNN

Effect of Reverberation Phenomena on Text- independent Speaker Recognition Based Deep Learning

A Comparative Study between Pitch Detection Techniques on Reverberant Speech Signals