Robust ENF Estimation Based on Harmonic Enhancement and Maximum Weight Clique

Hua, Guang; Liao, Han; Zhang, Haijian; Ye, Dengpan; Ma, Jiayi

doi:10.1109/tifs.2021.3099697

Cited by 15 publications

(23 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When recording audio in the electrical activity area, the ENF signal will be embedded into the audio [17]. The fluctuation of ENF in a specific area is stable and unique within a certain period [18], so ENF can be used to detect audio tampering [19,20]. Two existing methods for audio tampering detection using ENF include database comparison and consistency analysis.…”

Section: Detection Methods Based On Shallow Featuresmentioning

confidence: 99%

Shallow and deep feature fusion for digital audio tampering detection

Wang

Yang

Zeng

et al. 2022

EURASIP J. Adv. Signal Process.

View full text Add to dashboard Cite

Digital audio tampering detection can be used to verify the authenticity of digital audio. However, most current methods use standard electronic network frequency (ENF) databases for visual comparison analysis of ENF continuity of digital audio or perform feature extraction for classification by machine learning methods. ENF databases are usually tricky to obtain, visual methods have weak feature representation, and machine learning methods have more information loss in features, resulting in low detection accuracy. This paper proposes a fusion method of shallow and deep features to fully use ENF information by exploiting the complementary nature of features at different levels to more accurately describe the changes in inconsistency produced by tampering operations to raw digital audio. Firstly, the audio signal is band-pass filtered to obtain the ENF component. Then, the discrete Fourier transform (DFT) and Hilbert transform are performed to obtain the phase and instantaneous frequency of the ENF component. Secondly, the mean value of the sequence variation is used as the shallow feature; the feature matrix obtained by framing and reshaping of the ENF sequence is used as the input of the convolutional neural network; the characteristics of the fitted coefficients are obtained by curve fitting. Then, the local details of ENF are obtained from the feature matrix by the convolutional neural network, and the global information of ENF is obtained by fitting coefficient features through deep neural network (DNN). The depth features of ENF are composed of ENF global information and local information together. The shallow and deep features are fused using an attention mechanism to give greater weights to features useful for classification and suppress invalid features. Finally, the tampered audio is detected by downscaling and fitting with a DNN containing two fully connected layers, and classification is performed using a Softmax layer. The method achieves 97.03% accuracy on three classic databases: Carioca 1, Carioca 2, and New Spanish. In addition, we have achieved an accuracy of 88.31% on the newly constructed database GAUDI-DI. Experimental results show that the proposed method is superior to the state-of-the-art method.

show abstract

Section: Detection Methods Based On Shallow Featuresmentioning

confidence: 99%

Shallow and deep feature fusion for digital audio tampering detection

Wang

Yang

Zeng

et al. 2022

EURASIP J. Adv. Signal Process.

View full text Add to dashboard Cite

show abstract

“…In [79], the authors proposed a robust ENF extraction framework that included an improvement over the single-tone algorithm proposed in [78]. First, the authors introduced a harmonic robust filtering algorithm (HRFA) to extend singletone RFA to a multi-tone scenario.…”

Section: A Enf Extraction From Audio Recordingmentioning

confidence: 99%

“…Mitigating complex noise such as the multipath interference based on linear canonical transform [78], [79] Non-parametric, RFA, HRFA Enhancing the ENF signal relying on harmonic filtering in a low SNR situation.…”

Section: B Enf Extraction From Video Recordingmentioning

confidence: 99%

ENF Based Digital Multimedia Forensics: Survey, Application, Challenges and Future Work

Ngharamike,

Ang,

Seng

et al. 2023

IEEE Access

View full text Add to dashboard Cite

The electric network frequency (ENF) represents the transmission frequency of the electrical grid and fluctuates constantly around 50 Hz or 60 Hz, subject to the region. This constant fluctuation, caused by the continuous mismatch in power demand and supply, makes the ENF a unique signature, which can be utilized for several applications. According to studies, the ENF may be intrinsically implanted in audio recordings captured by digital audio recorders (e.g., microphone-based voice recorder) plugged into the mains supply or are situated close to sources of power and power transmission cable due to electromagnetic field interference originated from power source or acoustic hum and mechanical vibrations emitted by electrically operated devices such as regular household appliances. Recent studies further observed that video recordings made under an illumination source powered by main power can pick up the ENF signal. Following this discovery, several research efforts have been invested towards successful and accurate extraction of the ENF signal, and utilizing the ENF signal retrieved for several applications, including time stamp verification, audio/video authentication, location of recording estimation, power grid identification, estimation of camera read-out time, and video record synchronization. To the best of our knowledge, there has been no comprehensive survey on ENF-based multimedia forensics. Thus, in this paper, we present a comprehensive survey of studies conducted in this field, identifying several application specifics, current challenges, and future research directions.INDEX TERMS Electric network frequency (ENF), ENF detection, ENF estimation, audio and video forensic, ENF applications

show abstract

“…Guang Hua (Hua & Zhang, 2019) proposed a robust filtering algorithm (RFA) to enhance ENF signals in audio, making the extracted ENF signals more accurate. In (Hua et al, 2021), Guang Hua uses RFA to enhance each harmonic component of ENF and finally performs a weighted combination of harmonic components to obtain more accurate ENF estimation.…”

Section: Research On Tamper Detection Based On Electronic Network Fre...mentioning

confidence: 99%

Audio Tampering Forensics Based on Representation Learning of ENF Phase Sequence

Zeng

Yang

Wang

et al. 2022

International Journal of Digital Crime and Forensics

View full text Add to dashboard Cite

This paper proposes an audio tampering detection method based on the ENF phase and BI-LSTM network from the perspective of temporal feature representation learning. First, the ENF phase is obtained by discrete Fourier transform of ENF component in audio. Second, the ENF phase is divided into frames to obtain ENF phase sequence characterization, and each frame is represented as the change information of the ENF phase in a period. Then, the BI-LSTM neural network is used to train and output the state of each time step, and the difference information between real audio and tampered audio is obtained. Finally, these differences were fitted and dimensionally reduced by the fully connected network and classified by the Softmax classifier. Experimental results show that the performance of this method is better than the state-of-the-art approaches.

show abstract

Robust ENF Estimation Based on Harmonic Enhancement and Maximum Weight Clique

Cited by 15 publications

References 42 publications

Shallow and deep feature fusion for digital audio tampering detection

Shallow and deep feature fusion for digital audio tampering detection

ENF Based Digital Multimedia Forensics: Survey, Application, Challenges and Future Work

Audio Tampering Forensics Based on Representation Learning of ENF Phase Sequence

Contact Info

Product

Resources

About