Search

The prevalence of synthetic talking faces in both commercial and academic environments is increasing as the technology to generate them grows more powerful and available. While it has long been known that seeing the face of the talker improves human perception of speech-in-noise, recent studies have shown that synthetic talking faces generated by deep neural networks (DNNs) are also able to improve human perception of speech-in-noise. However, in previous studies the benefit provided by DNN synthetic faces was only about half that of real human talkers. We sought to determine whether synthetic talking faces generated by an alternative method would provide a greater perceptual benefit. The facial action coding system (FACS) is a comprehensive system for measuring visually discernible facial movements. Because the action units that comprise FACS are linked to specific muscle groups, synthetic talking faces generated by FACS might have greater verisimilitude than DNN synthetic faces which do not reference an explicit model of the facial musculature. We tested the ability of human observers to identity speech-in-noise accompanied by a blank screen; the real face of the talker; and synthetic talking faces generated either by DNN or FACS. We replicated previous findings of a large benefit for seeing the face of a real talker for speech-in-noise perception and a smaller benefit for DNN synthetic faces. FACS faces also improved perception, but only to the same degree as DNN faces. Analysis at the phoneme level showed that the performance of DNN and FACS faces was particularly poor for phonemes that involve interactions between the teeth and lips, such as /f/, /v/, and /th/. Inspection of single video frames revealed that the characteristic visual features for these phonemes were weak or absent in synthetic faces. Modeling the real vs. synthetic difference showed that increasing the realism of a few phonemes could substantially increase the overall perceptual benefit of synthetic faces. [ABSTRACT FROM AUTHOR]

In this paper we present a deep neural network modelling using Computational Fluid Dynamics (CFD) simulations data in order to optimize control of bioinspired morphing wings of a drone. Drones flight needs to consider variation in aerodynamic conditions that cannot all be optimized using a fixed aerodynamic profile. Nature solves this issue as birds are changing continuously the shape of their wings depending of the aerodynamic current requirements. One important issue for fixed wing drone is the landing as it is unable to control and most of the time consequences are some damages at the nose. An optimized shape of the wing at landing will avoid this situation. Another issue is that wings with a maximum surface are sensitive to stronger head winds; while wings with a small surface allowing the drone to fly faster. A wing with a morphing surface could adapt its aerial surface to optimize aerodynamic performance to specific flight situations. A morphing wing needs to be controlled in an optimized manner taking into account current aerodynamics parameters. Predicting optimized positions of the wing needs to consider (CFD) prior simulation parameters. The scenarios for flight require an important number of CFD simulation to address different conditions and geometric shapes. We compare in this paper neural network architecture suitable to predict wing shape according to current conditions. Deep neural network (DNN) is trained using data resulted out of CFD simulations to estimate flight conditions.

This paper proposes a new methodology for the automatic detection of magnetic disturbances from magnetic inertial measurement unit (MIMU) sensors based on deep learning. The proposed approach considers magnetometer data as input to a long short-term memory (LSTM) neural network and obtains a labeled time series output with the posterior probabilities of magnetic disturbance. We trained our algorithm on a data set that reproduces a wide range of magnetic perturbations and MIMU motions in a repeatable and reproducible way. The model was trained and tested using 15 folds, which considered independence in sensor, disturbance direction, and signal type. On average, the network can adequately detect the disturbances in 98 % of the cases, which represents a significant improvement over current threshold-based detection algorithms. [ABSTRACT FROM AUTHOR]

In this paper we present a deep neural network modelling using Computational Fluid Dynamics (CFD) simulations data in order to optimize control of bioinspired morphing wings of a drone. Drones flight needs to consider variation in aerodynamic conditions that cannot all be optimized using a fixed aerodynamic profile. Nature solves this issue as birds are changing continuously the shape of their wings depending of the aerodynamic current requirements. One important issue for fixed wing drone is the landing as it is unable to control and most of the time consequences are some damages at the nose. An optimized shape of the wing at landing will avoid this situation. Another issue is that wings with a maximum surface are sensitive to stronger head winds; while wings with a small surface allowing the drone to fly faster. A wing with a morphing surface could adapt its aerial surface to optimize aerodynamic performance to specific flight situations. A morphing wing needs to be controlled in an optimized manner taking into account current aerodynamics parameters. Predicting optimized positions of the wing needs to consider (CFD) prior simulation parameters. The scenarios for flight require an important number of CFD simulation to address different conditions and geometric shapes. We compare in this paper neural network architecture suitable to predict wing shape according to current conditions. Deep neural network (DNN) is trained using data resulted out of CFD simulations to estimate flight conditions. [ABSTRACT FROM AUTHOR]

This paper proposes a new methodology for the automatic detection of magnetic disturbances from magnetic inertial measurement unit (MIMU) sensors based on deep learning. The proposed approach considers magnetometer data as input to a long short-term memory (LSTM) neural network and obtains a labeled time series output with the posterior probabilities of magnetic disturbance. We trained our algorithm on a data set that reproduces a wide range of magnetic perturbations and MIMU motions in a repeatable and reproducible way. The model was trained and tested using 15 folds, which considered independence in sensor, disturbance direction, and signal type. On average, the network can adequately detect the disturbances in 98% of the cases, which represents a significant improvement over current threshold-based detection algorithms.

Summary: Nowadays, information security attacks are emerging as the main threat to network behavior. The deployment of sensor nodes in the open environment will be subjected to various kinds of DoS attacks and it needs to be analyzed. In this article, we considered two attacks namely, black hole attack and wormhole attack. The number of sensor nodes in the wireless network deployed in the network is taken as input data. Data packet drop and transmission latency are considered as two significant drawbacks of black hole and wormhole attack. Therefore, to conquer such limitations, the deep neural network based fuzzy imperialist competitive algorithm (DNN‐FICA) is developed which accurately detects the presence of attacker node in the wireless network. Here, the FICA approach is utilized to optimize the weights of DNN. The simulation is conducted using MATLAB software considering the performance metrics such as detection rate, packet delivery ratio, energy consumption, network lifetime, end‐to‐end delay, communication overhead, throughput, and residual energy. The comparative results are carried out and the result analysis proved that the proposed DNN‐FICA technique achieved greater overall performance than other compared methods. [ABSTRACT FROM AUTHOR]

The prevalence of synthetic talking faces in both commercial and academic environments is increasing as the technology to generate them grows more powerful and available. While it has long been known that seeing the face of the talker improves human perception of speech-in-noise, recent studies have shown that synthetic talking faces generated by deep neural networks (DNNs) are also able to improve human perception of speech-in-noise. However, in previous studies the benefit provided by DNN synthetic faces was only about half that of real human talkers. We sought to determine whether synthetic talking faces generated by an alternative method would provide a greater perceptual benefit. The facial action coding system (FACS) is a comprehensive system for measuring visually discernible facial movements. Because the action units that comprise FACS are linked to specific muscle groups, synthetic talking faces generated by FACS might have greater verisimilitude than DNN synthetic faces which do not reference an explicit model of the facial musculature. We tested the ability of human observers to identity speech-in-noise accompanied by a blank screen; the real face of the talker; and synthetic talking faces generated either by DNN or FACS. We replicated previous findings of a large benefit for seeing the face of a real talker for speech-in-noise perception and a smaller benefit for DNN synthetic faces. FACS faces also improved perception, but only to the same degree as DNN faces. Analysis at the phoneme level showed that the performance of DNN and FACS faces was particularly poor for phonemes that involve interactions between the teeth and lips, such as /f/, /v/, and /th/. Inspection of single video frames revealed that the characteristic visual features for these phonemes were weak or absent in synthetic faces. Modeling the real vs. synthetic difference showed that increasing the realism of a few phonemes could substantially increase the overall perceptual benefit of synthetic faces., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2024 Yu, Lado, Zhang, Magnotti and Beauchamp.)

Introduction: One attractive research area in the computer science field is soft biometrics., Aim: To Identify a person's gender from an iris image when such identification is related to security surveillance systems and forensics applications., Methods: In this paper, a robust iris gender-identification method based on a deep convolutional neural network is introduced. The proposed architecture segments the iris from a background image using the graph-cut segmentation technique. The proposed model contains 16 subsequent layers; three are convolutional layers for feature extraction with different convolution window sizes, followed by three fully connected layers for classification., Results: The original dataset consists of 3,000 images, 1,500 images for men and 1,500 images for women. The augmentation techniques adopted in this research overcome the overfitting problem and make the proposed architecture more robust and immune from simply memorizing the training data. In addition, the augmentation process not only increased the number of dataset images to 9,000 images for the training phase, 3,000 images for the testing phase and 3,000 images for the verification phase but also led to a significant improvement in testing accuracy, where the proposed architecture achieved 98.88%. A comparison is presented in which the testing accuracy of the proposed approach was compared with the testing accuracy of other related works using the same dataset., Conclusion: The proposed architecture outperformed the other related works in terms of testing accuracy., Competing Interests: There are no conflict of interest.