Camera-based Image Forgery Localization using Convolutional Neural Networks. Cozzolino, D. & Verdoliva, L. In 2018 26th European Signal Processing Conference (EUSIPCO), pages 1372-1376, Sep., 2018. Paper doi abstract bibtex Camera fingerprints are precious tools for a number of image forensics tasks. A well-known example is the photo response non-uniformity (PRNU) noise pattern, a powerful device fingerprint. Here, to address the image forgery localization problem, we rely on noiseprint, a recently proposed CNN-based camera model fingerprint. The CNN is trained to minimize the distance between same-model patches, and maximize the distance otherwise. As a result, the noiseprint accounts for model-related artifacts just like the PRNU accounts for device-related nonuniformities. However, unlike the PRNU, it is only mildly affected by residuals of high-level scene content. The experiments show that the proposed noiseprint-based forgery localization method improves over the PRNU-based reference.
@InProceedings{8553581,
author = {D. Cozzolino and L. Verdoliva},
booktitle = {2018 26th European Signal Processing Conference (EUSIPCO)},
title = {Camera-based Image Forgery Localization using Convolutional Neural Networks},
year = {2018},
pages = {1372-1376},
abstract = {Camera fingerprints are precious tools for a number of image forensics tasks. A well-known example is the photo response non-uniformity (PRNU) noise pattern, a powerful device fingerprint. Here, to address the image forgery localization problem, we rely on noiseprint, a recently proposed CNN-based camera model fingerprint. The CNN is trained to minimize the distance between same-model patches, and maximize the distance otherwise. As a result, the noiseprint accounts for model-related artifacts just like the PRNU accounts for device-related nonuniformities. However, unlike the PRNU, it is only mildly affected by residuals of high-level scene content. The experiments show that the proposed noiseprint-based forgery localization method improves over the PRNU-based reference.},
keywords = {cameras;feature extraction;feedforward neural nets;fingerprint identification;image forensics;image sensors;convolutional neural networks;camera fingerprints;image forensics tasks;photo response nonuniformity noise pattern;localization problem;CNN;noiseprint accounts;model-related artifacts;device-related nonuniformities;localization method;PRNU-based reference;camera-based image forgery localization;device fingerprint;PRNU;noiseprint-based forgery localization;high-level scene content;Cameras;Forgery;Training;Task analysis;Computational modeling;Forensics;Noise reduction;Image forensics;PRNU;convolutional neural networks},
doi = {10.23919/EUSIPCO.2018.8553581},
issn = {2076-1465},
month = {Sep.},
url = {https://www.eurasip.org/proceedings/eusipco/eusipco2018/papers/1570439441.pdf},
}
Downloads: 0
{"_id":"DWg9PKt4KTeWeFKAM","bibbaseid":"cozzolino-verdoliva-camerabasedimageforgerylocalizationusingconvolutionalneuralnetworks-2018","authorIDs":[],"author_short":["Cozzolino, D.","Verdoliva, L."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["D."],"propositions":[],"lastnames":["Cozzolino"],"suffixes":[]},{"firstnames":["L."],"propositions":[],"lastnames":["Verdoliva"],"suffixes":[]}],"booktitle":"2018 26th European Signal Processing Conference (EUSIPCO)","title":"Camera-based Image Forgery Localization using Convolutional Neural Networks","year":"2018","pages":"1372-1376","abstract":"Camera fingerprints are precious tools for a number of image forensics tasks. A well-known example is the photo response non-uniformity (PRNU) noise pattern, a powerful device fingerprint. Here, to address the image forgery localization problem, we rely on noiseprint, a recently proposed CNN-based camera model fingerprint. The CNN is trained to minimize the distance between same-model patches, and maximize the distance otherwise. As a result, the noiseprint accounts for model-related artifacts just like the PRNU accounts for device-related nonuniformities. However, unlike the PRNU, it is only mildly affected by residuals of high-level scene content. The experiments show that the proposed noiseprint-based forgery localization method improves over the PRNU-based reference.","keywords":"cameras;feature extraction;feedforward neural nets;fingerprint identification;image forensics;image sensors;convolutional neural networks;camera fingerprints;image forensics tasks;photo response nonuniformity noise pattern;localization problem;CNN;noiseprint accounts;model-related artifacts;device-related nonuniformities;localization method;PRNU-based reference;camera-based image forgery localization;device fingerprint;PRNU;noiseprint-based forgery localization;high-level scene content;Cameras;Forgery;Training;Task analysis;Computational modeling;Forensics;Noise reduction;Image forensics;PRNU;convolutional neural networks","doi":"10.23919/EUSIPCO.2018.8553581","issn":"2076-1465","month":"Sep.","url":"https://www.eurasip.org/proceedings/eusipco/eusipco2018/papers/1570439441.pdf","bibtex":"@InProceedings{8553581,\n author = {D. Cozzolino and L. Verdoliva},\n booktitle = {2018 26th European Signal Processing Conference (EUSIPCO)},\n title = {Camera-based Image Forgery Localization using Convolutional Neural Networks},\n year = {2018},\n pages = {1372-1376},\n abstract = {Camera fingerprints are precious tools for a number of image forensics tasks. A well-known example is the photo response non-uniformity (PRNU) noise pattern, a powerful device fingerprint. Here, to address the image forgery localization problem, we rely on noiseprint, a recently proposed CNN-based camera model fingerprint. The CNN is trained to minimize the distance between same-model patches, and maximize the distance otherwise. As a result, the noiseprint accounts for model-related artifacts just like the PRNU accounts for device-related nonuniformities. However, unlike the PRNU, it is only mildly affected by residuals of high-level scene content. The experiments show that the proposed noiseprint-based forgery localization method improves over the PRNU-based reference.},\n keywords = {cameras;feature extraction;feedforward neural nets;fingerprint identification;image forensics;image sensors;convolutional neural networks;camera fingerprints;image forensics tasks;photo response nonuniformity noise pattern;localization problem;CNN;noiseprint accounts;model-related artifacts;device-related nonuniformities;localization method;PRNU-based reference;camera-based image forgery localization;device fingerprint;PRNU;noiseprint-based forgery localization;high-level scene content;Cameras;Forgery;Training;Task analysis;Computational modeling;Forensics;Noise reduction;Image forensics;PRNU;convolutional neural networks},\n doi = {10.23919/EUSIPCO.2018.8553581},\n issn = {2076-1465},\n month = {Sep.},\n url = {https://www.eurasip.org/proceedings/eusipco/eusipco2018/papers/1570439441.pdf},\n}\n\n","author_short":["Cozzolino, D.","Verdoliva, L."],"key":"8553581","id":"8553581","bibbaseid":"cozzolino-verdoliva-camerabasedimageforgerylocalizationusingconvolutionalneuralnetworks-2018","role":"author","urls":{"Paper":"https://www.eurasip.org/proceedings/eusipco/eusipco2018/papers/1570439441.pdf"},"keyword":["cameras;feature extraction;feedforward neural nets;fingerprint identification;image forensics;image sensors;convolutional neural networks;camera fingerprints;image forensics tasks;photo response nonuniformity noise pattern;localization problem;CNN;noiseprint accounts;model-related artifacts;device-related nonuniformities;localization method;PRNU-based reference;camera-based image forgery localization;device fingerprint;PRNU;noiseprint-based forgery localization;high-level scene content;Cameras;Forgery;Training;Task analysis;Computational modeling;Forensics;Noise reduction;Image forensics;PRNU;convolutional neural networks"],"metadata":{"authorlinks":{}}},"bibtype":"inproceedings","biburl":"https://raw.githubusercontent.com/Roznn/EUSIPCO/main/eusipco2018url.bib","creationDate":"2021-02-13T15:38:40.500Z","downloads":0,"keywords":["cameras;feature extraction;feedforward neural nets;fingerprint identification;image forensics;image sensors;convolutional neural networks;camera fingerprints;image forensics tasks;photo response nonuniformity noise pattern;localization problem;cnn;noiseprint accounts;model-related artifacts;device-related nonuniformities;localization method;prnu-based reference;camera-based image forgery localization;device fingerprint;prnu;noiseprint-based forgery localization;high-level scene content;cameras;forgery;training;task analysis;computational modeling;forensics;noise reduction;image forensics;prnu;convolutional neural networks"],"search_terms":["camera","based","image","forgery","localization","using","convolutional","neural","networks","cozzolino","verdoliva"],"title":"Camera-based Image Forgery Localization using Convolutional Neural Networks","year":2018,"dataSources":["yiZioZximP7hphDpY","iuBeKSmaES2fHcEE9"]}