TR2008-638, Dartmouth College, Computer Science

Methods

Digital Image Ballistics from JPEG Quantization: A Followup Study
Hany Farid Department of Computer Science Dartmouth College Hanover NH 03755

Abstract

The lossy JPEG compression scheme employs a quantization table that controls the amount of compression achieved. Because different cameras typically employ different tables, a comparison of an images quantization scheme to a database of known cameras affords a simple technique for conrming or denying an images source. This report describes the analysis of quantization tables extracted from 1,000,000 images downloaded from Flickr.com.

Introduction

Since the JPEG image format has emerged as a virtual standard, most cameras encode images in this format. This lossy compression scheme allows camera manufacturers to balance the amount of compression and the quality of their images. This tradeoff is embodied in the JPEG quantization table. In earlier work [1], we showed that the quantization table is fairly distinct across various cameras, and can therefore be used to conrm or deny the source of an image (see also [2]). This initial study examined one image taken from one of 204 cameras, but suffered from not considering the different quantization tables that are employed with varying camera quality and resolution settings. In this report, we expand on our earlier work by considering a much larger and more diverse collection of images. 1
Using the Flickr API, 1,000,000 images were downloaded from Flickr.com, a popular photo-sharing website. Since we are interested in the JPEG quantization table employed by different cameras, it was necessary to eliminate any images that had been edited or altered by any photo-editing software. To this end, only images tagged as original by Flickr were downloaded. Images were then eliminated if they were not 3-channel color images, and further eliminated if they had incomplete metadata, inconsistent modication and original dates, or a software metadata tag suggesting that the image had been edited by a photoediting software. This ltering eliminated 557,045 images, leaving 442,955 images. The camera make, model, resolution, and JPEG quantization table were extracted from each of the 442,955 images metadata. In order to further eliminate possible edited or altered images, only those entries with ve or more images having the same paired make, model, resolution, and quantization table were retained. This ltering eliminated 105,329 images, leaving 337,626 images for the nal analysis. The remaining 337,626 images spanned 10,153 different distinct pairings of camera make, model, resolution and quantization table (accounting for 48 different camera manufacturers, and 859 different camera models). It is from these 10,153 entries that the distinctness of the paired resolution and quantization table was analyzed. Both the resolution and quantization table were combined in order to provide a more rened criteria for narrowing the search criteria to identify the source camera.

Results

The 10,153 entries described in the previous section fall into equivalence classes of size 1 to 178, where each class has identical resolution and JPEG quantization table. The average class size is 20.13, with a standard deviation of 35.7 (a class of size 1 means that the paired values are unique). The median class size is 5. Of the 10,153 entries, 2,704 entries (26.6%) have a unique paired resolution and quantization table, and 37.2% have at most two matches, and 44.1% have at most three matches.
Figure 1: The distribution of equivalence class sizes for paired resolution and JPEG quantization (a class size of 1 denotes a unique pairing).

Shown in Figure 1 is the distribution of the size of these equivalence classes, and shown in Appendix A are the cameras belonging to the same equivalence class. In the majority of cases, it is cameras from the same manufacturer that share the same quantization table, although in several situations there is a signicant amount of overlap between cameras of different manufacturers. If we ignore resolution and partition only on the quantization tables, then the 10,153 entries fall into equivalence classes of size 1 to 899, where each class has an identical quantization table. The average class size is 273.6, with a standard deviation of 276.1. The median class size is 254, and 517 (5.1%) have a unique quantization table, and 8.3% have at most two matches, and 10.4% entries have at most three matches.
and by the Institute for Security Technology Studies at Dartmouth College under grants from the Bureau of Justice Assistance (2005-DD-BX-1091) and the U.S. Department of Homeland Security (2006-CS-001-000001). Points of view or opinions in this document are those of the author and do not represent the ofcial position or policies of the U.S. Department of Justice, the U.S. Department of Homeland Security, or any other sponsor.

References

[1] H. Farid. Digital image ballistics from JPEG quantization. Technical Report TR2006-583, Department of Computer Science, Dartmouth College, 2006. [2] J.D. Kornblum. Using JPEG quantization tables to identify imagery processed by software. Digital Investigation, 5:S21S25, 2008.

Discussion

While the JPEG quantization table is clearly not unique, it (paired with the image resolution) is reasonably effective at narrowing the source of an image to a single camera make and model or to a small set of possible cameras.

Appendix A

Each entry in the rst table corresponds to a camera with unique resolution and quantization table. Each entry in the second table corresponds to an equivalence class of identical resolution and Acknowledgments quantization table. The size of the equivalence class is in the rst column, and the camera make This work was supported by a gift from Adobe and model and resolution are in the second colSystems, Inc., a gift from Microsoft, Inc., a grant from the National Science Foundation (CNS-0708209), umn. 2

Olympus E-410 (2681 3575) Olympus E-500 (1296 1728) Olympus E-510 (1368 1824) Olympus E-520 (1540 2047) Olympus FE280,X820,C520 (2448 3264) Olympus SP510UZ (2304 3072) Olympus u1010,S1010 (2736 3648) Olympus u1030SW,S1030SW (480 640) Olympus u30D,S410D,u410D (1536 2048) Olympus u710,S710 (1536 2048) Olympus u760,S760 (2304 3072) Olympus u770SW,S770SW (480 640) Olympus u790SW,S790SW (960 1280) Olympus u850SW,S850SW (1536 2048) Olympus uD600,S600 (1200 1600) Olympus X-3,C-60Z (2112 2816) Panasonic DMC-FS5 (2736 3648) Panasonic DMC-FX01 (2112 2816) Panasonic DMC-FX07 (480 640) Panasonic DMC-FX10 (2112 2816) Panasonic DMC-FX30 (1367 1823) Panasonic DMC-FX30 (576 768) Panasonic DMC-FX37 (2056 3648) Panasonic DMC-FX50 (2048 3072) Panasonic DMC-FZ18 (1224 1632) Panasonic DMC-FZ18 (450 600) Panasonic DMC-FZ18 (819 1024) Panasonic DMC-FZ20 (1920 2560) Panasonic DMC-FZ28 (1536 2048) Panasonic DMC-FZ30 (1536 2048) Panasonic DMC-FZ30 (2160 3248) Panasonic DMC-FZ50 (2160 3248) Panasonic DMC-FZ5 (1200 1600) Panasonic DMC-FZ7 (1200 1600) Panasonic DMC-FZ7 (1880 2816) Panasonic DMC-FZ8 (1050 1400) Panasonic DMC-FZ8 (2304 3072) Panasonic DMC-LC20 (1200 1600) Panasonic DMC-LS2 (960 1280) Panasonic DMC-LS60 (1584 2816) Panasonic DMC-LX2 (1080 1920) Panasonic DMC-LX2 (2376 4224) Panasonic DMC-LX2 (808 1420) Panasonic DMC-LX3 (1335 2000) Panasonic DMC-LX3 (2520 3776) Panasonic DMC-LZ3 (1920 2560) Panasonic DMC-LZ6 (1728 3072) Panasonic DMC-LZ8 (1536 2048) Panasonic DMC-TZ1 (857 1200) Panasonic DMC-TZ3 (1200 1600) Panasonic DMC-TZ3 (1872 3328) Panasonic DMC-TZ3 (413 550) Panasonic DMC-TZ4 (1920 2560) Panasonic DMC-TZ5 (1036 1382) Panasonic DMC-TZ5 (1440 2560) Panasonic DMC-TZ5 (2592 3456) Pentax *ist DS (2000 3008) Pentax K100D (736 1109) Pentax K10D (1249 1867) Pentax K10D (2000 3000) Pentax K10D (450 671) Pentax K10D (857 1280) Pentax K20D (1276 1920) Pentax Optio 330 (1536 2048) Pentax Optio 750Z (1539 2048) Pentax Optio MX (1536 2048) Pentax Optio S4 (1728 2304) Pentax Optio S6 (2112 2816) Pentax Optio SV (1920 2560) Research In Motion BlackBerry 8100 Series (480 640) Ricoh GR DIGITAL 2 (2432 3648) Ricoh GX200 (3000 4000) Samsung Digimax 301 (480 640) Samsung HMX20 (2160 2880) Samsung <KENOX S1050/S1050> (2736 3648) Samsung <NV3, VLUU NV3> (2304 3072) Samsung U-CA 4 Digital Camera (1704 2272) Samsung VLUU L83T/ L83T (480 640) Samsung VLUU NV 7, NV 7 (375 500) Sanyo E6 (326 448) Sanyo E6 (332 448) Sanyo VPC-S500 (960 1280) Sony CD MAVICA (1704 2272) Sony CYBERSHOT (1704 2272) Sony CYBERSHOT (578 800) Sony CYBERSHOT U (1224 1632) Sony DCR-PC100 (294 300) Sony DCR-SR200 (1296 2304) Sony DSC-F828 (2448 3264) Sony DSC-H1 (1728 2592) Sony DSC-H2 (528 704) Sony DSC-H50 (576 1024) Sony DSC-H5 (1875 2500) Sony DSC-H7 (2176 3264) Sony DSC-N2 (2736 3648) Sony DSC-P200 (2048 3072) Sony DSC-P32 (1360 2048)
Olympus E-420 (1920 2560) Olympus E-500 (2448 3264) Olympus E-510 (1641 2188) Olympus E-520 (2736 3648) Olympus m:robe MR-500i (960 1280) Olympus SP510UZ (769 1024) Olympus u1030SW,S1030SW (1920 2560) Olympus u10D,S300D,u300D (1536 2048) Olympus u30D,S410D,u410D (1704 2272) Olympus u710,S710 (2304 3072) Olympus u770SW,S770SW (1536 2048) Olympus u790SW,S790SW (1536 2048) Olympus u820,S820 (2448 3264) Olympus u850SW,S850SW (1920 2560) Olympus uD600,S600 (2112 2816) Panasonic DMC-FS5 (2048 3072) Panasonic DMC-FX01 (1200 1600) Panasonic DMC-FX01 (480 640) Panasonic DMC-FX100 (2248 4000) Panasonic DMC-FX12 (2304 3072) Panasonic DMC-FX30 (1920 2560) Panasonic DMC-FX3 (1880 2816) Panasonic DMC-FX500 (2736 3648) Panasonic DMC-FX55 (1080 1920) Panasonic DMC-FZ18 (1712 2560) Panasonic DMC-FZ18 (817 1024) Panasonic DMC-FZ18 (820 1024) Panasonic DMC-FZ20 (240 320) Panasonic DMC-FZ28 (2736 3648) Panasonic DMC-FZ30 (1712 2560) Panasonic DMC-FZ3 (1512 2016) Panasonic DMC-FZ50 (2400 3600) Panasonic DMC-FZ5 (1920 2560) Panasonic DMC-FZ7 (1360 2048) Panasonic DMC-FZ7 (2112 2816) Panasonic DMC-FZ8 (1080 1920) Panasonic DMC-FZ8 (535 800) Panasonic DMC-LC5 (840 1120) Panasonic DMC-LS60 (1050 1400) Panasonic DMC-LS70 (1536 2048) Panasonic DMC-LX2 (1152 2048) Panasonic DMC-LX2 (576 1024) Panasonic DMC-LX2 (852 1280) Panasonic DMC-LX3 (1768 2656) Panasonic DMC-LX3 (854 1280) Panasonic DMC-LZ5 (2112 2816) Panasonic DMC-LZ7 (1536 2048) Panasonic DMC-TZ1 (1920 2560) Panasonic DMC-TZ1 (858 1200) Panasonic DMC-TZ3 (1440 2560) Panasonic DMC-TZ3 (2144 3216) Panasonic DMC-TZ3 (960 1280) Panasonic DMC-TZ4 (2176 3264) Panasonic DMC-TZ5 (1296 1728) Panasonic DMC-TZ5 (2088 3712) Pentacon DCZ 6.8 (596 795) Pentax *ist DS2 (2000 3008) Pentax K100D (788 1050) Pentax K10D (1308 1948) Pentax K10D (2000 3008) Pentax K10D (671 1000) Pentax K10D (967 1280) Pentax K20D (3104 4672) Pentax Optio 330GS (1536 2048) Pentax Optio A10 (2448 3264) Pentax Optio S10 (1944 2592) Pentax Optio S55 (1536 2048) Pentax Optio S7 (1728 2304) Research In Motion BlackBerry 8100 Series (1024 1280) Research In Motion BlackBerry 8310 (1200 1600) Ricoh GR DIGITAL 2 (2736 2736) Ricoh R8 (2832 3744) Samsung Digimax 301 (768 1024) Samsung i70 (2304 3072) Samsung <KENOX S630/S630> (2112 2816) Samsung S85 (2448 3264) Samsung VLUU L100, M100/L100, M100 (1875 2500) Samsung VLUU NV10, NV10 (2016 2688) Sanyo C5 (2760 3680) Sanyo E6 (329 448) Sanyo E6 (334 448) SeaLife DC800 (480 640) Sony CYBERSHOT (1200 1600) Sony CYBERSHOT (1920 2560) Sony CYBERSHOT (600 800) Sony DCR-DVD505 (1728 2304) Sony DCR-PC110 (480 640) Sony DCR-SR65 (864 1152) Sony DSC-F828 (759 1012) Sony DSC-H1 (576 768) Sony DSC-H3 (1080 1920) Sony DSC-H50 (596 795) Sony DSC-H5 (336 504) Sony DSC-H7 (371 495) Sony DSC-P100 (1728 2592) Sony DSC-P200 (384 512) Sony DSC-P43 (960 1280)