Real-time facial expression recognition using smoothed deep neural network ensemble

Benamara, Nadir Kamel; Val-Calvo, Mikel; Álvarez-Sánchez, José R.; Díaz‐Morcillo, Alejandro; Vicente, José Manuel Ferrández; Fernández-Jover, Eduardo; Stambouli, Tarik Boudghene

doi:10.3233/ica-200643

Cited by 29 publications

(11 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Industry 4.0 [1] represents an innovative technological approach, where Cyber-Physical Systems [2], Artificial Intelligence [35,72], automatic decision systems [23], optimization solutions [18], exponential technologies [74], robotics [73] and circular business models are building the productive fabric [3,4].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Lightweight encryption for short-range wireless biometric authentication systems in Industry 4.0

Bordel

2022

ICA

View full text Add to dashboard Cite

Most recent solutions for users’ authentication in Industry 4.0 scenarios are based on unique biological characteristics that are captured from users and recognized using artificial intelligence and machine learning technologies. These biometric applications tend to be computationally heavy, so to monitor users in an unobtrusive manner, sensing and processing modules are physically separated and connected through point-to-point wireless communication technologies. However, in this approach, sensors are very resource constrained, and common cryptographic techniques to protect private users’ information while traveling in the radio channel cannot be implemented because their computational cost. Thus, new security solutions for those biometric authentication systems in their short-range wireless communications are needed. Therefore, in this paper, we propose a new cryptographic approach addressing this scenario. The proposed solution employs lightweight operations to create a secure symmetric encryption solution. This cipher includes a pseudo-random number generator based, also, on simple computationally low-cost operations in order to create the secret key. In order to preserve and provide good security properties, the key generation and the encryption processes are fed with a chaotic number sequence obtained through the numerical integration of a new four-order hyperchaotic dynamic. An experimental analysis and a performance evaluation are provided in the experimental section, showing the good behavior of the described solution.

show abstract

Section: Introductionmentioning

confidence: 99%

“…Hereinafter, the term "lightweight" refers to mechanisms with a very reduced computational cost or power [6]. In our work, we are looking for an algorithm requiring a very limited computation time.…”

Section: Introductionmentioning

confidence: 99%

Lightweight encryption for short-range wireless biometric authentication systems in Industry 4.0

Bordel

2022

ICA

View full text Add to dashboard Cite

show abstract

“…When the angle between the direction of railway tracks and the direction of positioning anchors becomes relatively small or even close to zero, the traditional anchor‐based representation fails to represent the railway tracks effectively. Thus, like the concept of network ensemble methods (Alam et al., 2019; Benamara et al., 2020; McCoy et al., 2022), this paper proposes an anchor‐adaptive dual‐branch railway track representation method for aerial images.…”

Section: Methodsmentioning

confidence: 99%

Anchor‐adaptive railway track detection from unmanned aerial vehicle images

Tong

Jia

Geng

et al. 2023

Computer aided Civil Eng

View full text Add to dashboard Cite

Autonomous railway inspection with unmanned aerial vehicles (UAVs) has huge advantages over traditional inspection methods. As a prerequisite for UAV‐based autonomous following of railway lines, it is quite essential to develop intelligent railway track detection algorithms. However, there are no existing algorithms currently that can efficiently adapt to the demand for the various forms and changing inclination angles of railway tracks in the UAV aerial images. To address the challenge, this paper proposes a novel anchor‐adaptive railway track detection network (ARTNet), which constructs a dual‐branch architecture based on projection length discrimination to realize full‐angle railway track detection for the UAV aerial images taken from arbitrary viewing angles. Considering the potential capacity imbalance of the two branches that can be caused by the uneven distribution of railway tracks in the dataset, a balanced transpose co‐training strategy is proposed to train the two branches coordinately. Moreover, an extra customized transposed consistency loss is designed to guide the training of the network without increasing any computational complexity. A set of experiments have been conducted to verify the feasibility and superiority of the ARTNet. It is demonstrated that our approach can effectively realize full‐angle railway track detection and outperform other popular algorithms greatly in terms of both detection accuracy and reasoning efficiency. ARTNet can achieve a mean F1 of 76.12 and run at a speed of 50 more frames per second.

show abstract

“…By the use of fully convolutional layers in NNs it is possible to extract e.g important and recurring parts of images. In [16] Benamara et al for example applied convolutional NNs to extract human emotions based on facial images. Autoencoders like in [17][18][19] are able to learn important features of the input data by first encoding the input data to a lower dimensional space and then decoding it to reconstruct the input.…”

Section: Related Workmentioning

confidence: 99%

Pattern discovery in time series using autoencoder in comparison to nonlearning approaches

Noering

Schroeder

Jonas

et al. 2021

ICA

View full text Add to dashboard Cite

In technical systems the analysis of similar situations is a promising technique to gain information about the system’s state, its health or wearing. Very often, situations cannot be defined but need to be discovered as recurrent patterns within time series data of the system under consideration. This paper addresses the assessment of different approaches to discover frequent variable-length patterns in time series. Because of the success of artificial neural networks (NN) in various research fields, a special issue of this work is the applicability of NNs to the problem of pattern discovery in time series. Therefore we applied and adapted a Convolutional Autoencoder and compared it to classical nonlearning approaches based on Dynamic Time Warping, based on time series discretization as well as based on the Matrix Profile. These nonlearning approaches have also been adapted, to fulfill our requirements like the discovery of potentially time scaled patterns from noisy time series. We showed the performance (quality, computing time, effort of parametrization) of those approaches in an extensive test with synthetic data sets. Additionally the transferability to other data sets is tested by using real life vehicle data. We demonstrated the ability of Convolutional Autoencoders to discover patterns in an unsupervised way. Furthermore the tests showed, that the Autoencoder is able to discover patterns with a similar quality like classical nonlearning approaches.

show abstract

Real-time facial expression recognition using smoothed deep neural network ensemble

Cited by 29 publications

References 33 publications

Lightweight encryption for short-range wireless biometric authentication systems in Industry 4.0

Lightweight encryption for short-range wireless biometric authentication systems in Industry 4.0

Anchor‐adaptive railway track detection from unmanned aerial vehicle images

Pattern discovery in time series using autoencoder in comparison to nonlearning approaches

Contact Info

Product

Resources

About