Papers by Athanassios Skodras
A New Forgery Image Dataset and its Subjective Evaluation
2023 IEEE IAS Global Conference on Emerging Technologies (GlobConET)
A Novel Finger Vein Recognition System Based on Enhanced Maximum Curvature Points
2018 IEEE 13th Image, Video, and Multidimensional Signal Processing Workshop (IVMSP)
Finger vein recognition is a biometric method of authentication that offers high security, effici... more Finger vein recognition is a biometric method of authentication that offers high security, efficiency and stability. In this paper we propose a new finger vein recognition system that utilizes the Enhanced Maximum Curvature Points (EMC) technique for finger vein pattern extraction and introduces a new pre-processing stage. In addition, it combines two matching methods, leading to better recognition performance in terms of EER, FAR, FRR and recognition rate than other methods. We present the experimental results obtained by applying our system on the databases SDUMLA-HMT, Tsingua, FV-USM and HKPU and compare them with similar approaches applied on these databases.
Compact FPGA architectures for the two-band fast discrete Hartley transform
Microprocessors and Microsystems
Abstract The discrete Hartley transform is a real valued transform similar to the complex Fourier... more Abstract The discrete Hartley transform is a real valued transform similar to the complex Fourier transform that finds numerous applications in a variety of fields including pattern recognition and signal and image processing. In this paper, we propose and study two compact and versatile hardware architectures for the computation of the 8-point, 16-point and 32-point Two-Band Fast Discrete Hartley Transform. These highly modular architectures have a symmetric and regular structure consisting of two blocks, a multiplication block and an addition/subtraction block. The first architecture utilizes 8 multipliers and 16 adders/subtractors, achieving a maximum clock frequency of 95 MHz. The second architecture utilizes only 4 multipliers and 8 adders/subtractors, achieving a maximum clock frequency of 100 MHz; however it requires additional multiplexers and more clock cycles (from 1 to 58 clock cycles depends on the points) for the computation. As a result, the proposed hardware architectures constitute an efficient choice for area-restricted applications such as embedded or pervasive computing systems.
Pruning of the two-dimensional fast cosine transform algorithm
Video Surveillance Authentication: Real-Time ENF Signal Hiding at the Edge
2023 24th International Conference on Digital Signal Processing (DSP)
Due to the significance of the visual information exchanged in Internet of Video Things (IoVT) ne... more Due to the significance of the visual information exchanged in Internet of Video Things (IoVT) networks, attackers are constantly launching new attacks and attempt to exploit new vulnerabilities. One of the most common and difficult-to- prevent attacks on the Visual Layer is the Frame Duplication Attack (FDA). Recently, two techniques were proposed for FDA detection at the edge by using the embedded Electrical Network Frequency (ENF) signals in an effort to surpass limitations of conventional passive methods. In this paper, a Real-Time ENF signal hiding technique at the edge is proposed. Our motivation is to examine the possibility of authenticating the surveillance feed by hiding the ENF signal. Experiments are conducted, including an extensive performance comparison between the proposed and reference encoder, a feasibility study for the proposed encoder’s integration to a Raspberry Pi for video streaming purposes and finally the implementation of a proof-of-concept prototype. According to the findings, the proposed approach provides real- time FDA detection at reduced computational complexity and hardware requirements, thus rendering this method appropriate for applications at the edge.
Learned Image Compression with Wavelet Preprocessing for Low Bit Rates
2023 24th International Conference on Digital Signal Processing (DSP)
Could Human Gaze Augment Detectors of Synthetic Images?
2023 24th International Conference on Digital Signal Processing (DSP)
A Hilbert Curve Based Representation of sEMG Signals for Gesture Recognition
2019 International Conference on Systems, Signals and Image Processing (IWSSIP)
Deep learning (DL) has transformed the field of data analysis by dramatically improving the state... more Deep learning (DL) has transformed the field of data analysis by dramatically improving the state of the art in various classification and prediction tasks, especially in the area of computer vision. In biomedical engineering, a lot of new work is directed towards surface electromyography (sEMG) based gesture recognition, often addressed as an image classification problem using Convolutional Neural Networks (CNN). In this paper, we utilize the Hilbert space-filling curve for the generation of image representations of sEMG signals that are then classified by CNN. The proposed method is evaluated on different network architectures and yields a classification improvement of more than 3%.
Proportional Myoelectric Control in a Virtual Reality Environment
2022 IEEE 14th Image, Video, and Multidimensional Signal Processing Workshop (IVMSP)
On the Use of Deeper CNNs in Hand Gesture Recognition Based on sEMG Signals
2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA), 2019
In the past few years, a great interest for the classification of hand gestures with Deep Learnin... more In the past few years, a great interest for the classification of hand gestures with Deep Learning methods based on surface electromyography (sEMG) signals has been developed in the scientific community. In line with latest works in the field, the objective of our work is the construction of a novel Convolutional Neural Network architecture, for the classification of hand-gestures. Our model, while avoiding overfitting, did not perform significantly better compared to a much shallower network. The results suggest that the lack of diversity in the sEMG recordings between certain hand-gestures limits the performance of the model. In addition, the classification accuracy on a database we developed using a commercial device (Myo Armband) was substantially higher (approximately 24%) than a similar benchmark dataset recorded with the same device.
JPEG2000, the new standard for still image coding, Quantization, user defined wavelets, arbitrary... more JPEG2000, the new standard for still image coding, Quantization, user defined wavelets, arbitrary wavelet provides a new framework and an integrated toolbox to better decompositions, general scaling-based ROI coding, and address increasing needs for compression. It offers a wide advanced error resilience schemes (Fig. 1). It is actually a range of functionalities such as lossless and lossy coding, toolbox with technologies useful for various specialized embedded lossy to lossless coding, progression by resolution applications. Part 3 defines motion JPEG2000 (MJ2 or and quality, high compression efficiency, error resilience and MJP2) and is based on Part 1 of JPEG2000. MJ2 will be region-of-interest (ROI) coding. Comparative results have
Training Makers to Build the Internet of Things on an Arduino (Using a Remote Lab Facility and an MOOC)
The Internet of Things for Education, 2021
Over the past years, Deep Learning methods have shown promising results to a wide range of resear... more Over the past years, Deep Learning methods have shown promising results to a wide range of research fields including image classification and natural language processing. Their increased success rates have drawn the attention of many researchers from various domains. This chapter investigates the application of Deep Learning methods to the problem of electromyography-based gesture recognition. A signal processing pipeline based on Deep Learning is presented through examples taken from the literature, whereas the details of state-of-the-art neural network architectures are discussed. In addition, this chapter illustrates a few ways adopted from image classification tasks that visualize what the neural network learns. Finally, new approaches are proposed and evaluated with publicly available datasets. Keywords— sEMG, gesture recognition, Deep Learning, signal processing
On the use of the Kinect Sensor in Rehabilitation
2018 Tenth International Conference on Quality of Multimedia Experience (QoMEX), 2018
Point clouds are one of the most promising technologies for 3D content representation. In this pa... more Point clouds are one of the most promising technologies for 3D content representation. In this paper, we describe a study on quality assessment of point clouds, degraded by octreebased compression on different levels. The test contents were displayed using Screened Poisson surface reconstruction, without including any textural information, and they were rated by subjects in a passive way, using a 2D image sequence. Subjective evaluations were performed in five independent laboratories in different countries, with the inter-laboratory correlation analysis showing no statistical differences, despite the different equipment employed. Benchmarking results reveal that the state-of-the-art point cloud objective metrics are not able to accurately predict the expected visual quality of such test contents. Moreover, the subjective scores collected from this experiment were found to be poorly correlated with subjective scores obtained from another test involving visualization of raw point clouds. These results suggest the need for further investigations on adequate point cloud representations and objective quality assessment tools.
Transfer Learning in sEMG-based Gesture Recognition
2021 12th International Conference on Information, Intelligence, Systems & Applications (IISA), 2021
The latest advancements in the field of deep learning and biomedical engineering have allowed for... more The latest advancements in the field of deep learning and biomedical engineering have allowed for the development of myoelectric interfaces based on deep neural networks. A longstanding problem of these interfaces is that the models cannot easily be applied to new users due to the high variability and stochastic nature of the electromyography signals. Further training a new model for every new subject requires the collection of large volumes of data. Therefore, this work proposes a transfer learning (TL) scheme which allows reusing the knowledge of a pre-existing model for a new user. Firstly, a convolutional neural network (CNN) is trained on an initial dataset using the data of multiple subjects. Then, the weights of this model are fine-tuned for a new target subject. The approach is evaluated on the Ninapro datasets DB2 and DB7. The experimentation included three different CNN models and eight preprocessing alternatives. The results showed that the success of the TL method depends on how the data are preprocessed. Specifically, the biggest accuracy improvement (+5.14%) is achieved when only the first 20% of the signal duration is used.
Improved Gesture Recognition Based on sEMG Signals and TCN
ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2019
In recent years, the successful application of Deep Learning methods to classification problems h... more In recent years, the successful application of Deep Learning methods to classification problems has had a huge impact in many domains. In biomedical engineering, the problem of gesture recognition based on electromyography is often addressed as an image classification problem using Convolutional Neural Networks. In this paper, we approach electromyography-based hand gesture recognition as a sequence classification problem using Temporal Convolutional Networks. The proposed network yields an improvement in gesture recognition of almost 5% to the state of the art reported in the literature, whereas the analysis helps in understanding the limitations of the model and exploring new ways to improve its performance.
Proceedings of the 5th International Conference on Physiological Computing Systems, 2018
In recent years, Deep Learning methods have been successfully applied to a wide range of image an... more In recent years, Deep Learning methods have been successfully applied to a wide range of image and speech recognition problems highly impacting other research fields. As a result, new works in biomedical engineering are directed towards the application of these methods to electromyography-based gesture recognition. In this paper, we present a brief overview of Deep Learning methods for electromyography-based hand gesture recognition along with an analysis of a modified simple model based on Convolutional Neural Networks. The proposed network yields a 3% improvement on the classification accuracy of the basic model, whereas the analysis helps in understanding the limitations of the model and exploring new ways to improve the performance.
Sensors, 2022
In recent years, the successful application of Deep Learning methods to classification problems h... more In recent years, the successful application of Deep Learning methods to classification problems has had a huge impact in many domains. (1) Background: In biomedical engineering, the problem of gesture recognition based on electromyography is often addressed as an image classification problem using Convolutional Neural Networks. Recently, a specific class of these models called Temporal Convolutional Networks (TCNs) has been successfully applied to this task. (2) Methods: In this paper, we approach electromyography-based hand gesture recognition as a sequence classification problem using TCNs. Specifically, we investigate the real-time behavior of our previous TCN model by performing a simulation experiment on a recorded sEMG dataset. (3) Results: The proposed network trained with data augmentation yields a small improvement in accuracy compared to our existing model. However, the classification accuracy is decreased in the real-time evaluation, showing that the proposed TCN architec...
2002 14th International Conference on Digital Signal Processing Proceedings. DSP 2002 (Cat. No.02TH8628)
Uploads
Papers by Athanassios Skodras