PRNU-based Image Classification of Origin Social Network with CNN. Caldelli, R., Amerini, I., & Li, C. T. In 2018 26th European Signal Processing Conference (EUSIPCO), pages 1357-1361, Sep., 2018. Paper doi abstract bibtex A huge amount of images are continuously shared on social networks (SNs) daily and, in most of cases, it is very difficult to reliably establish the SN of provenance of an image when it is recovered from a hard disk, a SD card or a smartphone memory. During an investigation, it could be crucial to be able to distinguish images coming directly from a photo-camera with respect to those downloaded from a social network and possibly, in this last circumstance, determining which is the SN among a defined group. It is well known that each SN leaves peculiar traces on each content during the upload-download process; such traces can be exploited to make image classification. In this work, the idea is to use the PRNU, embedded in every acquired images, as the “carrier” of the particular SN traces which diversely modulate the PRNU. We demonstrate, in this paper, that SN-modulated noise residual can be adopted as a feature to detect the social network of origin by means of a trained convolutional neural network (CNN).
@InProceedings{8553160,
author = {R. Caldelli and I. Amerini and C. T. Li},
booktitle = {2018 26th European Signal Processing Conference (EUSIPCO)},
title = {PRNU-based Image Classification of Origin Social Network with CNN},
year = {2018},
pages = {1357-1361},
abstract = {A huge amount of images are continuously shared on social networks (SNs) daily and, in most of cases, it is very difficult to reliably establish the SN of provenance of an image when it is recovered from a hard disk, a SD card or a smartphone memory. During an investigation, it could be crucial to be able to distinguish images coming directly from a photo-camera with respect to those downloaded from a social network and possibly, in this last circumstance, determining which is the SN among a defined group. It is well known that each SN leaves peculiar traces on each content during the upload-download process; such traces can be exploited to make image classification. In this work, the idea is to use the PRNU, embedded in every acquired images, as the “carrier” of the particular SN traces which diversely modulate the PRNU. We demonstrate, in this paper, that SN-modulated noise residual can be adopted as a feature to detect the social network of origin by means of a trained convolutional neural network (CNN).},
keywords = {cameras;convolution;feedforward neural nets;image classification;object detection;smart phones;social networking (online);SN-modulated noise residual;CNN;PRNU-based image classification;hard disk;upload-download process;convolutional neural network;social network;smartphone memory;images acquisition;Cameras;Transform coding;Facebook;Training;Feature extraction;Europe},
doi = {10.23919/EUSIPCO.2018.8553160},
issn = {2076-1465},
month = {Sep.},
url = {https://www.eurasip.org/proceedings/eusipco/eusipco2018/papers/1570437218.pdf},
}
Downloads: 0
{"_id":"2tThQEYDHe8wKGnkP","bibbaseid":"caldelli-amerini-li-prnubasedimageclassificationoforiginsocialnetworkwithcnn-2018","authorIDs":[],"author_short":["Caldelli, R.","Amerini, I.","Li, C. T."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["R."],"propositions":[],"lastnames":["Caldelli"],"suffixes":[]},{"firstnames":["I."],"propositions":[],"lastnames":["Amerini"],"suffixes":[]},{"firstnames":["C.","T."],"propositions":[],"lastnames":["Li"],"suffixes":[]}],"booktitle":"2018 26th European Signal Processing Conference (EUSIPCO)","title":"PRNU-based Image Classification of Origin Social Network with CNN","year":"2018","pages":"1357-1361","abstract":"A huge amount of images are continuously shared on social networks (SNs) daily and, in most of cases, it is very difficult to reliably establish the SN of provenance of an image when it is recovered from a hard disk, a SD card or a smartphone memory. During an investigation, it could be crucial to be able to distinguish images coming directly from a photo-camera with respect to those downloaded from a social network and possibly, in this last circumstance, determining which is the SN among a defined group. It is well known that each SN leaves peculiar traces on each content during the upload-download process; such traces can be exploited to make image classification. In this work, the idea is to use the PRNU, embedded in every acquired images, as the “carrier” of the particular SN traces which diversely modulate the PRNU. We demonstrate, in this paper, that SN-modulated noise residual can be adopted as a feature to detect the social network of origin by means of a trained convolutional neural network (CNN).","keywords":"cameras;convolution;feedforward neural nets;image classification;object detection;smart phones;social networking (online);SN-modulated noise residual;CNN;PRNU-based image classification;hard disk;upload-download process;convolutional neural network;social network;smartphone memory;images acquisition;Cameras;Transform coding;Facebook;Training;Feature extraction;Europe","doi":"10.23919/EUSIPCO.2018.8553160","issn":"2076-1465","month":"Sep.","url":"https://www.eurasip.org/proceedings/eusipco/eusipco2018/papers/1570437218.pdf","bibtex":"@InProceedings{8553160,\n author = {R. Caldelli and I. Amerini and C. T. Li},\n booktitle = {2018 26th European Signal Processing Conference (EUSIPCO)},\n title = {PRNU-based Image Classification of Origin Social Network with CNN},\n year = {2018},\n pages = {1357-1361},\n abstract = {A huge amount of images are continuously shared on social networks (SNs) daily and, in most of cases, it is very difficult to reliably establish the SN of provenance of an image when it is recovered from a hard disk, a SD card or a smartphone memory. During an investigation, it could be crucial to be able to distinguish images coming directly from a photo-camera with respect to those downloaded from a social network and possibly, in this last circumstance, determining which is the SN among a defined group. It is well known that each SN leaves peculiar traces on each content during the upload-download process; such traces can be exploited to make image classification. In this work, the idea is to use the PRNU, embedded in every acquired images, as the “carrier” of the particular SN traces which diversely modulate the PRNU. We demonstrate, in this paper, that SN-modulated noise residual can be adopted as a feature to detect the social network of origin by means of a trained convolutional neural network (CNN).},\n keywords = {cameras;convolution;feedforward neural nets;image classification;object detection;smart phones;social networking (online);SN-modulated noise residual;CNN;PRNU-based image classification;hard disk;upload-download process;convolutional neural network;social network;smartphone memory;images acquisition;Cameras;Transform coding;Facebook;Training;Feature extraction;Europe},\n doi = {10.23919/EUSIPCO.2018.8553160},\n issn = {2076-1465},\n month = {Sep.},\n url = {https://www.eurasip.org/proceedings/eusipco/eusipco2018/papers/1570437218.pdf},\n}\n\n","author_short":["Caldelli, R.","Amerini, I.","Li, C. T."],"key":"8553160","id":"8553160","bibbaseid":"caldelli-amerini-li-prnubasedimageclassificationoforiginsocialnetworkwithcnn-2018","role":"author","urls":{"Paper":"https://www.eurasip.org/proceedings/eusipco/eusipco2018/papers/1570437218.pdf"},"keyword":["cameras;convolution;feedforward neural nets;image classification;object detection;smart phones;social networking (online);SN-modulated noise residual;CNN;PRNU-based image classification;hard disk;upload-download process;convolutional neural network;social network;smartphone memory;images acquisition;Cameras;Transform coding;Facebook;Training;Feature extraction;Europe"],"metadata":{"authorlinks":{}},"downloads":0},"bibtype":"inproceedings","biburl":"https://raw.githubusercontent.com/Roznn/EUSIPCO/main/eusipco2018url.bib","creationDate":"2020-03-18T03:31:20.646Z","downloads":0,"keywords":["cameras;convolution;feedforward neural nets;image classification;object detection;smart phones;social networking (online);sn-modulated noise residual;cnn;prnu-based image classification;hard disk;upload-download process;convolutional neural network;social network;smartphone memory;images acquisition;cameras;transform coding;facebook;training;feature extraction;europe"],"search_terms":["prnu","based","image","classification","origin","social","network","cnn","caldelli","amerini","li"],"title":"PRNU-based Image Classification of Origin Social Network with CNN","year":2018,"dataSources":["yiZioZximP7hphDpY","iuBeKSmaES2fHcEE9"]}