Smartphone Camera Identification from Low-Mid Frequency DCT Coefficients of Dark Images
Camera sensor identification can have numerous forensics and authentication applications. In this work, we follow an identification methodology for smartphone camera sensors using properties of the Dark Signal Nonuniformity (DSNU) in the collected images. This requires taking dark pictures, which th...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-08-01
|
| Series: | Entropy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1099-4300/24/8/1158 |
| _version_ | 1797410456295440384 |
|---|---|
| author | Adriana Berdich Bogdan Groza |
| author_facet | Adriana Berdich Bogdan Groza |
| author_sort | Adriana Berdich |
| collection | DOAJ |
| description | Camera sensor identification can have numerous forensics and authentication applications. In this work, we follow an identification methodology for smartphone camera sensors using properties of the Dark Signal Nonuniformity (DSNU) in the collected images. This requires taking dark pictures, which the users can easily do by keeping the phone against their palm, and has already been proposed by various works. From such pictures, we extract low and mid frequency AC coefficients from the DCT (Discrete Cosine Transform) and classify the data with the help of machine learning techniques. Traditional algorithms such as KNN (K-Nearest Neighbor) give reasonable results in the classification, but we obtain the best results with a wide neural network, which, despite its simplicity, surpassed even a more complex network architecture that we tried. Our analysis showed that the blue channel provided the best separation, which is in contrast to previous works that have recommended the green channel for its higher encoding power. |
| first_indexed | 2024-03-09T04:29:11Z |
| format | Article |
| id | doaj.art-d778497692bd47e9bfbc81b36273e2ca |
| institution | Directory Open Access Journal |
| issn | 1099-4300 |
| language | English |
| last_indexed | 2024-03-09T04:29:11Z |
| publishDate | 2022-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Entropy |
| spelling | doaj.art-d778497692bd47e9bfbc81b36273e2ca2023-12-03T13:37:27ZengMDPI AGEntropy1099-43002022-08-01248115810.3390/e24081158Smartphone Camera Identification from Low-Mid Frequency DCT Coefficients of Dark ImagesAdriana Berdich0Bogdan Groza1Faculty of Automatics and Computers, Politehnica University of Timisoara, 300223 Timisoara, RomaniaFaculty of Automatics and Computers, Politehnica University of Timisoara, 300223 Timisoara, RomaniaCamera sensor identification can have numerous forensics and authentication applications. In this work, we follow an identification methodology for smartphone camera sensors using properties of the Dark Signal Nonuniformity (DSNU) in the collected images. This requires taking dark pictures, which the users can easily do by keeping the phone against their palm, and has already been proposed by various works. From such pictures, we extract low and mid frequency AC coefficients from the DCT (Discrete Cosine Transform) and classify the data with the help of machine learning techniques. Traditional algorithms such as KNN (K-Nearest Neighbor) give reasonable results in the classification, but we obtain the best results with a wide neural network, which, despite its simplicity, surpassed even a more complex network architecture that we tried. Our analysis showed that the blue channel provided the best separation, which is in contrast to previous works that have recommended the green channel for its higher encoding power.https://www.mdpi.com/1099-4300/24/8/1158smartphonecamera sensorfingerprintingDSNUAC coefficientsmachine learning |
| spellingShingle | Adriana Berdich Bogdan Groza Smartphone Camera Identification from Low-Mid Frequency DCT Coefficients of Dark Images Entropy smartphone camera sensor fingerprinting DSNU AC coefficients machine learning |
| title | Smartphone Camera Identification from Low-Mid Frequency DCT Coefficients of Dark Images |
| title_full | Smartphone Camera Identification from Low-Mid Frequency DCT Coefficients of Dark Images |
| title_fullStr | Smartphone Camera Identification from Low-Mid Frequency DCT Coefficients of Dark Images |
| title_full_unstemmed | Smartphone Camera Identification from Low-Mid Frequency DCT Coefficients of Dark Images |
| title_short | Smartphone Camera Identification from Low-Mid Frequency DCT Coefficients of Dark Images |
| title_sort | smartphone camera identification from low mid frequency dct coefficients of dark images |
| topic | smartphone camera sensor fingerprinting DSNU AC coefficients machine learning |
| url | https://www.mdpi.com/1099-4300/24/8/1158 |
| work_keys_str_mv | AT adrianaberdich smartphonecameraidentificationfromlowmidfrequencydctcoefficientsofdarkimages AT bogdangroza smartphonecameraidentificationfromlowmidfrequencydctcoefficientsofdarkimages |