Christos Korgialas

A novel framework, called InterGridNet, is introduced, leveraging a shallow RawNet model for geolocation classification of Electric Network Frequency (ENF) signatures in the SP Cup 2016 dataset. During data preparation, recordings are sorted into audio and power groups based on inherent characteristics, further divided into 50 Hz and 60 Hz groups via spectrogram analysis. Residual blocks within the classification model extract frame-level embeddings, aiding decision-making through softmax activation. The topology and the hyperparameters of the shallow RawNet are optimized using a Neural Architecture Search. The overall accuracy of InterGridNet in the test recordings is 92%, indicating its effectiveness against the state-of-the-art methods tested in the SP Cup 2016. These findings underscore InterGridNet's effectiveness in accurately classifying audio recordings from diverse power grids, advancing state-of-the-art geolocation estimation methods.

Source Device Identification (SDI) is pivotal in multimedia forensics, as it entails the recognition of the device that captured a specific image or video. This paper introduces an innovative SDI method using log-Mel spectrograms from video audio, employing an optimized ResNet-based model enhanced with Neural Architecture Search and integrated with Gradient-weighted Class Activation Mapping (Grad-CAM) for insights into influential spectrogram regions. A strong emphasis on high-frequency components within the audio data is observed, allowing band-pass filtering to the input spectrograms to selectively retain high-frequency information, resulting in the highest classification accuracy of camera models. Experiments conducted on the VISION dataset, comprising data from 35 different devices, demonstrate the effectiveness of the proposed method in achieving accurate and interpretable SDI, marking the first application of explainable Artificial Intelligence (xAI) techniques resorting to Grad-CAM in this context. Furthermore, a bootstrap analysis is performed to evaluate the classification performance impact of the proposed methodology with and without the integration of Grad-CAM explanations. By assessing the Grad-CAM-driven method, featuring band-pass filtered log-Mel spectrograms against the state-of-the-art approaches, the high accuracy in SDI is illustrated.

The Electric Network Frequency (ENF) serves as a unique signature inherent to power distribution systems. Here, a novel approach for power grid classification is developed, leveraging ENF. Spectrograms are generated from audio and power recordings across different grids, revealing distinctive ENF patterns that aid in grid classification through a fusion of classifiers. Four traditional machine learning classifiers plus a Convolutional Neural Network (CNN), optimized using Neural Architecture Search, are developed for One-vs-All classification. This process generates numerous predictions per sample, which are then compiled and used to train a shallow multi-label neural network specifically designed to model the fusion process, ultimately leading to the conclusive class prediction for each sample. Experimental findings reveal that both validation and testing accuracy outperform those of current state-of-the-art classifiers, underlining the effectiveness and robustness of the proposed methodology.

The Electric Network Frequency (ENF) serves as a simple means to verify the authenticity of audio recordings. ENF variations contain crucial information, acting as a distinctive “fingerprint” when electronic devices are connected or located near power mains. A novel framework for ENF estimation is proposed. This approach alternates between the Least Absolute Deviation (LAD) regression for determining regression weights and objective function minimization with respect to frequency, adapting them within the context of the $\ell _{1}$ norm or the sum of $\ell _{1}$ norms of the approximation error. This framework is a direct consequence of Laplacian distributed noise. Goodness-of-fit tests are reported, indicating that the Laplacian noise hypothesis is more appropriate than the hypothesis of Gaussian noise in the benchmark ENF-WHU dataset. Extensive evaluation using audio recordings from the aforementioned dataset demonstrates the exceptional performance of the proposed framework outperforming state-of-the-art ENF estimation schemes. These findings provide compelling evidence for the efficacy of the proposed ENF estimation schemes as reliable prerequisites for detecting audio forgeries.

A robust regression technique known as Huber regression is incorporated into the Electric Network Frequency (ENF) detection task. This novel framework is based on the assumption of a mixture noise model, which combines Gaussian and Laplacian noise for ENF detection in short-length audio recordings. The effectiveness of the proposed ENF detector is assessed through accuracy calculations and the analysis of the Receiver Operating Characteristic curve with respect to the Area Under the Curve. Real-world benchmark data from the ENF-WHU dataset are utilized for this evaluation. The experimental results indicate that integrating the Huber regression method leads to a significant enhancement in ENF detection for short-length audio recordings, outperforming the performance of existing state-of-the-art techniques.

Electric Network Frequency (ENF) is a fingerprint in multi-media forensics applications. ENF is a weak signal that is difficult to be detected. This difficulty stems from the existence of colored wide-sense stationary Gaussian noise in ENF as well as due to many unknown random parameters. However, several ENF detectors have been proposed, motivating the related research. In this paper, a novel Least Absolute Deviations-based ENF detector is proposed that is coined as LAD-Likelihood Ratio Test (LAD-LRT). The performance of the LAD-LRT detector is thoroughly analyzed concerning test statistic distribution and threshold selection. The aim is to develop a detector that detects ENF more accurately in short-length recordings than the state-of-the-art Least-Squares (LS)-LRT and naive-LRT detectors. Thorough evaluation using benchmark audio recordings demonstrate the effectiveness of the proposed detector.

This paper deals with Generative Adversarial Networks (GANs) applied to face aging. An explainable face aging framework is proposed that builds on a well-known face aging approach, namely the Conditional Adversarial Autoencoder (CAAE). The proposed framework, namely, xAI-CAAE, couples CAAE with explainable Artificial Intelligence (xAI) methods, such as Saliency maps or Shapley additive explanations, to provide corrective feedback from the discriminator to the generator. xAI-guided training aims to supplement this feedback with explanations that provide a “reason” for the discriminator’s decision. Moreover, Local Interpretable Model-agnostic Explanations (LIME) are leveraged to provide explanations for the face areas that most influence the decision of a pre-trained age classifier. To the best of our knowledge, xAI methods are utilized in the context of face aging for the first time. A thorough qualitative and quantitative evaluation demonstrates that the incorporation of the xAI systems contributed significantly to the generation of more realistic age-progressed and regressed images.