sift-forensic

所属分类:matlab编程
开发工具:matlab
文件大小:1596KB
下载次数:76
上传日期:2014-07-02 00:32:03
上 传 者Mariam1
说明:  detection of copy move forgery using sift

文件列表:
sift-forensic (0, 2012-05-10)
sift-forensic\examples (0, 2012-05-10)
sift-forensic\lib (0, 2012-05-07)
sift-forensic\dataset (0, 2012-05-10)
sift-forensic\lib\sift (0, 2010-12-08)
sift-forensic\lib\sift\src (0, 2012-05-07)
sift-forensic\lib\sift\docs (0, 2012-05-07)
sift-forensic\lib\sift\bin (0, 2012-05-07)
sift-forensic\lib\sift\lib (0, 2012-05-07)
sift-forensic\lib\sift\include (0, 2010-12-08)
sift-forensic\lib\sift\docs\html (0, 2012-05-07)
sift-forensic\import_sift.m (1708, 2012-05-09)
sift-forensic\run_F2000_experiment.m (1659, 2012-05-10)
sift-forensic\hnormalise.m (1031, 2012-05-07)
sift-forensic\homography2d.m (2408, 2011-07-26)
sift-forensic\match_features.m (3666, 2012-05-09)
sift-forensic\iscolinear.m (2313, 2012-05-07)
sift-forensic\normalise2dpts2.m (2563, 2012-05-07)
sift-forensic\show_inliers.m (375, 2012-05-08)
sift-forensic\ransac.m (9926, 2011-07-26)
sift-forensic\ransacfithomography2.m (5374, 2012-05-07)
sift-forensic\run_F220_experiment.m (1657, 2012-05-10)
sift-forensic\process_image.m (3642, 2012-05-10)
sift-forensic\tampering_localization.m (2131, 2012-05-08)
sift-forensic\vgg_condition_2d.m (256, 2012-05-07)
sift-forensic\vgg_get_nonhomg.m (375, 2011-07-26)
sift-forensic\vgg_Haffine_from_x_MLE.m (1473, 2012-05-07)
sift-forensic\examples\tampered4.jpg (387806, 2011-03-08)
sift-forensic\examples\tampered2.jpg (95322, 2012-05-07)
sift-forensic\examples\tampered3.jpg (35974, 2012-05-07)
sift-forensic\examples\tampered1.jpg (70026, 2011-07-26)
sift-forensic\examples\tampered2.sift (1750657, 2012-05-08)
sift-forensic\dataset\results_F220_F2000.txt (381, 2012-05-10)
sift-forensic\lib\sift\LICENSE (3206, 2010-06-20)
sift-forensic\lib\sift\mainpage (2191, 2010-06-20)
sift-forensic\lib\sift\Doxyfile (45594, 2010-06-20)
sift-forensic\lib\sift\beaver_xform.png (19484, 2010-06-20)
sift-forensic\lib\sift\THANKS (651, 2010-12-08)
sift-forensic\lib\sift\beaver.sift (42774, 2010-06-20)
... ...

COPY-MOVE FORGERY DETECTION --- This code was developed by I. Amerini, L. Ballan, G. Serra at Media Integration and Communication Center, University of Florence (Italy). If you use this code please cite the paper: I. Amerini, L. Ballan, R. Caldelli, A. Del Bimbo, and G. Serra, "A SIFT-based forensic method for copy-move attack detection and transformation recovery", IEEE Trans. on Information Forensics and Security, vol. 6, iss. 3, pp. 1099-1110, 2011. --- May 8, 2012 - version 1.0 1. Introduction --- This package contains the Matlab implementation of the copy-move detection approach presented in Amerini et al., TIFS 2011. Our code use several public functions and libraries developed by other authors; regarding these files, for any problem or license information, please refer to the respective authors. In particular, SIFT features are extracted using the Rob Hess implementation (http://blogs.oregonstate.edu/hess/code/sift/). Anyway, the code works also loading features extraced with other implementations (such as the original one by D. Lowe). 2. Main functions --- - process_image(imagefile, metric, thc, min_cluster_pts, plotimg, extractsift) The function runs Copy-Move-Detection on a single image. For example run from the Matlab prompt: >> process_image('examples/tampered1.jpg', 'ward', 3, 4, 1) You can also import a file containing sift descriptors if you have pre-computed features: >> process_image('examples/tampered2.jpg', 'ward', 3, 4, 1, 'examples/tampered2.sift') - run_F220_experiment and run_F2000_experiment These two scripts can be used to replicate the experiments reported in our paper; you have only to download these two datasets and unzip them into the 'dataset' directory. You can find this data at the URL: http://www.micc.unifi.it/ballan/research/image-forensics/ Please note that you will probably obtain very similar (but not the same) results with respect to those reported in our paper since, in that case, we applied a 4-fold cross validation procedure that is not implemented in these scripts. The expected results are reported in a text file in the dataset directory. 3. Contacts --- Irene Amerini (irene.amerini@unifi.it) Lamberto Ballan (ballan@dsi.unifi.it) Giuseppe Serra (serra@dsi.unifi.it)

近期下载者

相关文件


收藏者