 Forensics of image re-sampling (such as image resizing) is an important issue,which can be used for tampering detection, steganography, etc.  Most of.

Slides:



Advertisements
Similar presentations
QR Code Recognition Based On Image Processing
Advertisements

Road-Sign Detection and Recognition Based on Support Vector Machines Saturnino, Sergio et al. Yunjia Man ECG 782 Dr. Brendan.
Feature-Based Steganalysis for JPEG images and its applications for future design of steganographic schemes. - Jessica Fridrich Submitted by: Praveena.
1 Image Authentication by Detecting Traces of Demosaicing June 23, 2008 Andrew C. Gallagher 1,2 Tsuhan Chen 1 Carnegie Mellon University 1 Eastman Kodak.
Digital Watermarking for Telltale Tamper Proofing and Authentication Deepa Kundur, Dimitrios Hatzinakos Presentation by Kin-chung Wong.
1 A robust detection algorithm for copy- move forgery in digital images Source: Forensic Science International, Volume 214, Issues 1–3, 10 January 2012.
Camera Model Identification Based on the Characteristics of CFA and Interpolation Shang Gao 1, Guanshuo Xu 2, Rui-Min Hu 1,*
Maximizing Strength of Digital Watermarks Using Neural Network Presented by Bin-Cheng Tzeng 5/ Kenneth J.Davis; Kayvan Najarian International Conference.
A Comprehensive Study on Third Order Statistical Features for Image Splicing Detection Xudong Zhao, Shilin Wang, Shenghong Li and Jianhua Li Shanghai Jiao.
Face Recognition & Biometric Systems, 2005/2006 Face recognition process.
Chih-Hsing Lin, Jia-Shiuan Tsai, and Ching-Te Chiu
Efficient Bit Allocation and CTU level Rate Control for HEVC Picture Coding Symposium, 2013, IEEE Junjun Si, Siwei Ma, Wen Gao Insitute of Digital Media,
ECE643 DIGITAL IMAGE PROCESSING Steganalysis versus Splicing detection Paper by: Yun Q. Shi, Chunhua Chen, Guorong Xuan and Wei Su By: Nehal Patel Siddharth.
Detecting Image Region Duplication Using SIFT Features March 16, ICASSP 2010 Dallas, TX Xunyu Pan and Siwei Lyu Computer Science Department University.
Student: Hsu-Yung Cheng Advisor: Jenq-Neng Hwang, Professor
1 A Unified Rate-Distortion Analysis Framework for Transform Coding Student : Ho-Chang Wu Student : Ho-Chang Wu Advisor : Prof. David W. Lin Advisor :
A New Rate-Complexity-QP Algorithm for HEVC Intra-Picture Rate Control LING TIAN, YIMIN ZHOU, AND XIAOJUN CAO 2014 INTERNATIONAL CONFERENCE ON COMPUTING,
EI San Jose, CA Slide No. 1 Nearest-neighbor and Bilinear Resampling Factor Estimation to Detect Blockiness or Blurriness of an Image* Ariawan Suwendi.
A Review on: Spread Spectrum Watermarking Techniques
Digital Watermarking Parag Agarwal
Median Filtering Detection Using Edge Based Prediction Matrix The 10th IWDW, Atlantic City, New Jersey, USA 23~26 October 2011 School of Information Science.
Retaliating Anti-forensics of JPEG Image Compression Based On the Noise Level Estimation PROPOSAL SPRING 2015 ADVISOR: Dr. K.R.Rao Presented by, Komandla.
Introduction to Visible Watermarking IPR Course: TA Lecture 2002/12/18 NTU CSIE R105.
Huijuan Yang, Alex C. Kot, IEEE Fellow IEEE Transactions on Multimedia, Vol. 9, No. 3, Apr Multimedia Security Final Project R 葉容瑜 R
Video Tracking Using Learned Hierarchical Features
Blind Pattern Matching Attack on Watermark Systems D. Kirovski and F. A. P. Petitcolas IEEE Transactions on Signal Processing, VOL. 51, NO. 4, April 2003.
DCT-Domain Watermarking Chiou-Ting Hsu and Ja-Ling Wu, "Hidden digital watermarks in images," IEEE Trans. On Image Processing, vol. 8, No. 1, January 1999.
1 Security and Robustness Enhancement for Image Data Hiding Authors: Ning Liu, Palak Amin, and K. P. Subbalakshmi, Senior Member, IEEE IEEE TRANSACTIONS.
A Novel steganographic method for JPEG images by Vasiliy Sachnev - Introduction  JPEG compression  Steganography - Block based steganography method (F5)
Competence Centre on Information Extraction and Image Understanding for Earth Observation Prof. Dr. Mihai Datcu SATELLITE IMAGE ARTIFACTS DETECTION BASED.
Eyes detection in compressed domain using classification Eng. Alexandru POPA Technical University of Cluj-Napoca Faculty.
Retaliating Anti-forensics of JPEG Image Compression Based On the Noise Level Estimation FINAL PRESENTATION SPRING 2015 ADVISOR: Dr. K.R.Rao Presented.
Introduction to Steganalysis Schemes Multimedia Security.
Exposing Digital Forgeries in Color Filter Array Interpolated Images By Alin C. Popescu and Hany Farid Presenting - Anat Kaspi.
Human Re-identification by Matching Compositional Template with Cluster Sampling Yuanlu Xu 1, Liang Lin 1, Wei-Shi Zheng 1, Xiaobai Liu 2 Abstract This.
An Improved Method Of Content Based Image Watermarking Arvind Kumar Parthasarathy and Subhash Kak 黃阡廷 2008/12/3.
Human Detection Method Combining HOG and Cumulative Sum based Binary Pattern Jong Gook Ko', Jin Woo Choi', So Hee Park', Jang Hee You', ' Electronics and.
Secure Spread Spectrum Watermarking for Multimedia Young K Hwang.
1 Watermarking Scheme Capable of Resisting Sensitivity Attack IEEE signal processing letters, vol. 14, no. 2, February. 2007, pp Xinpeng Zhang.
Blind image data hiding based on self reference Source : Pattern Recognition Letters, Vol. 25, Aug. 2004, pp Authors: Yulin Wang and Alan Pearmain.
Program Homework Implementation of the Improved Spread Spectrum Watermarking System.
By Pushpita Biswas Under the guidance of Prof. S.Mukhopadhyay and Prof. P.K.Biswas.
SuperResolution (SR): “Classical” SR (model-based) Linear interpolation (with post-processing) Edge-directed interpolation (simple idea) Example-based.
Learning Photographic Global Tonal Adjustment with a Database of Input / Output Image Pairs.
WCPM 1 Chang-Tsun Li Department of Computer Science University of Warwick UK Image Clustering Based on Camera Fingerprints.
Performance Measurement of Image Processing Algorithms By Dr. Rajeev Srivastava ITBHU, Varanasi.
MMC LAB Secure Spread Spectrum Watermarking for Multimedia KAIST MMC LAB Seung jin Ryu 1MMC LAB.
WLD: A Robust Local Image Descriptor Jie Chen, Shiguang Shan, Chu He, Guoying Zhao, Matti Pietikäinen, Xilin Chen, Wen Gao 报告人:蒲薇榄.
Date of download: 6/3/2016 Copyright © 2016 SPIE. All rights reserved. A framework for improved pedestrian detection performance through blind image distortion.
Digital Image Watermarking using Hybrid DWT -FFT Technique with Different Attacks Presented By: Mahendra Kumar Faculty at UCE, RTU, Kota (Raj.) India Director,
Watermarking Scheme Capable of Resisting Sensitivity Attack
Face Detection EE368 Final Project Group 14 Ping Hsin Lee
Reference Ingemar J. Cox, Joe Kilian, F. Thomson Leighton, and Talal Shamoon, "Secure Spread Spectrum Watermarking for Multimedia," IEEE Trans. on Image.
ABSTRACT FACE RECOGNITION RESULTS
Reversible Data Hiding in JPEG Images using Ordered Embedding
Cheng-Ming Huang, Wen-Hung Liao Department of Computer Science
Image Segmentation Techniques
Exposing Digital Forgeries by Detecting Traces of Resampling Alin C
Deep Learning Hierarchical Representations for Image Steganalysis
Aline Martin ECE738 Project – Spring 2005
Reduction of blocking artifacts in DCT-coded images
Improved joint reversible data hiding in encrypted images
Author: Minoru Kuribayashi, Hatsukazu Tanaka
Support vector machine-based text detection in digital video
New Framework for Reversible Data Hiding in Encrypted Domain
Source: J. Vis. Commun. Image R. 31 (2015) 64–74
Source: IEEE Transactions on Circuits and Systems,
I-Chuan Chang Bor-Wen Hsu and Chi Sung Laih
Detecting Digital Forgeries using Blind Noise Estimation
Reversible data hiding in encrypted binary images by pixel prediction
Presentation transcript:

 Forensics of image re-sampling (such as image resizing) is an important issue,which can be used for tampering detection, steganography, etc.  Most of the algorithms employed the periodical artifacts, which are caused by regular sampling, to detect image re- sampling.  Kirchner et al. [1] proposed two anti-forensic methods by removing the periodic artifacts using irregular sampling, and defeated all periodicity-based methods.  Attack based on geometric distortion with edge modulation (attack 1)  The dual-path approach (attack 2)  Our work  Our work aims at distinguishing the forged image from the original image. It can be considered as patches of periodicity- based methods.  We find the anti-forensic scheme leaves behind new clues. The last step of the anti-forensic scheme is interpolation, which changes the linear relationships among neighboring pixels..  To capture such relationships, the partial autocorrelation coefficients of neighboring pixels are taken as the feature. [1] M. Kirchner and R. Böhme, “Hiding traces of re-sampling in digital images”, IEEE Trans. Inf. Forensics Security, vol. 3, no.4, pp. 582–592, Dec Countering Anti-Forensics of Image Re-sampling Anjie Peng; Hui Zeng; Xiaodan Lin; Xiangui Kang School of Information Sci. & Tech., Sun Yat-sen Univ., China. Contact: Abstract Image re-sampling leaves behind periodical artifacts which are used as fingerprints for the forensics. A knowledgeable anti-forensic method erases such artifacts by irregular sampling. We observe that the irregular sampling followed by interpolation causes changes in local linear correlations, and propose a novel method to detect the anti-forensic method of re-sampling via partial autocorrelation coefficients. Experimental results on a large set of images show that the proposed method could effectively detect the anti-forensics of re-sampling with a low dimensional feature set. Introduction Experimental Results The Proposed Algorithm  Baseline test: classifying anti-forensics images from original images. Both two kinds are un-compressed.  Generalization ability test Tips: The classifier is trained on UCID database, while tested on BOSSbase. The test set has original images and anti-forensic forged images.  The goal of the proposed method aims at distinguishing the anti-forensic forged image from the original image.  Interpolation used in the anti-forensic scheme changes the relations among neighboring pixels, and thus leaves behind clues for detecting.  The partial autocorrelation is used to reveal the correlations among neighboring pixels. Given a sequence z, unlike the commonly used autocorrelation  k, the partial autocorrelation  kk is the autocorrelation between z t and z t+k with the linear dependency of z t+1 through to z t+k-1 removed. The estimation of  kk is as (3).  In this paper, we first analyze the processing chains of anti-forensics for image re-sampling, and then propose a novel feature set with low dimension to detect the anti- forensics of re- sampling.  Experimental results of baseline test and generalization ability test show that the proposed method could effectively detect the anti-forensics. Conclusion Tips:  SVM is used as a machine learning tool. All classifiers are trained and tested on the UCID database. Half images are used for training,other half are for testing.  Each test set has 669 original images and 669 anti-forensic forged images which are uniformly used scaling factors in [0.6, 2] with interval 0.1.  Pe means the minimal average decision error.  Robustness against JPEG compression Tips: The classifier is trained and tested on JPEG compressed UCID database. The anti-forensic. Forged image is generated by the following set (attack 1,  =0.4, kernel=‘bicubic’).  The partial autocorrelation feature (PAF) is extracted from both horizontal and vertical directions. Coefficients of two directions are averaged to get the PAF feature. Coefficients are extracted from image I and its 2nd-laplacian difference domain (denoted by D). The dimension of PAF is 12. PAF=[PAF I,PAF D ] Table 1. Pe ( % ) of the proposed scheme for 18 kinds of forged image. The second column shows the parameters used in attack,  : attack strength, kernel: interpolation algorithm.