bughht/CMFD

Copy-Move Forgery Detection and Localization: Course project for media information security

CMFD

Copy-Move Forgery Detection and Localization This is a course project for media information security

Introduction

Copy-Move Forgery Detection (CMFD) is a technique to detect and localize copy-move forgery in images. The goal of this project is to implement multiple CMFD algorithms in python and evaluate the performance.

We design a framework to evaluate the performance of the algorithm. The framework is based on PyTorch and can be easily extended to other algorithms.

Besides, we also implement a baseline algorithm (SIFT) and enhance it with patched self-adaptive methods to improve the performance.

Group Info

• Haotian Hong
• Zhenyu Jin

Phenomenon: Talk is cheap, show me the code.

Dataset

MICC-F220: this dataset is composed by 220 images; 110 are tampered and 110 originals.

Pre-requisites

• python>=3.7
• opencv-python
• numpy
• sklearn
• torch
• pandas

Installation

git clone https://github.com/bughht/CMFD.git
cd CMFD
wget http://lci.micc.unifi.it/labd/cmfd/MICC-F220.zip
unzip MICC-F220.zip
pip install -r Requirements.txt

Usage

Run the baseline

python AlgoTest.py -a SIFT_Methods

Design your own algorithm

Make sure your algorithm is written in the format below:

Filename: MyAlgorithm.py

class MyAlgorithm:
    def __init__(self, **kwargs):
        # initialize your algorithm

    def predict(self, img):
        # detect copy-move forgery in the image
        # return the classification result (0 or 1)

Then run the following command:

python AlgoTest.py -a MyAlgorithm

or

python AlgoTest.py --algorithm MyAlgorithm

Baseline: SIFT

The baseline is a sift-based algorithm implemented in Python. With current parameters, the evaluation of this algorithm on MICC-F220 is shown below.

Accuracy: 81.36% Precision: 76.74% Recall: 90.00% F1 Score: 82.85%

Enhancement: Patch-SIFT

• Principle: Split images into patches and adapt the parameters of SIFT (sigma) to the smoothness of the patch.
• Algorithm:
  • Split the image into patches
  • For each patch, calculate the smoothness of the patch (using the variance of the Laplacian)
  • For each patch, adapt the parameters of SIFT (sigma) to the smoothness of the patch (using linear model) and apply SIFT to the patch
  • Apply Brute-Force Matching to the image
  • Evaluate the performance of the algorithm

Accuracy: 86.82% Precision: 86.49% Recall: 87.27% F1 Score: 86.88%

Experiment Results

We've tested the following algorithms on MICC-F220 dataset based on our framework:

• Patch-SIFT
• SIFT
• ORB
• FAST

Patch-SIFT

Patch-SIFT	precision	recall	f1-score	support
No Copy-Move	0.87	0.86	0.87	110
Copy-Move	0.86	0.87	0.87	110

• Accuracy:86.82% Precision:86.49% Recall:87.27%
• Confusion Matrix

SIFT

SIFT	precision	recall	f1-score	support
No Copy-Move	0.88	0.73	0.80	110
Copy-Move	0.77	0.90	0.83	110

• Accuracy:81.36% Precision:76.74% Recall:90.00% F1 Score:82.85%
• Confusion-Matrix:

ORB

ORB	precision	recall	f1-score	support
No Copy-Move	0.65	0.75	0.69	110
Copy-Move	0.70	0.60	0.65	110

• Accuracy:67.27% Precision:70.21% Recall:60.00% F1 Score:64.71%
• Confusion-Matrix:

Project Goals Checkbox

• Implement feature-point-based algorithms
  • Key Points Extraction
    • SIFT
    • ORB
    • FAST
    • Harris Corner
  • Feature Descriptor
    • SIFT feature
    • ORB feature
• Implement matching algorithms
  • Brute-force matching
  • Fann matching
• Design a model performance evaluation framework
  • Torch Dataset and DataLoader wrapper
  • Model performance evaluation
• Enhance one of the algorithm tested above

Contribution

We are welcome to any contribution to this project. If you are interested in this project, please contact us.

References

Expand all

[1]	FADL S M, SEMARY N A. Robust copy--move forgery revealing in digital images using polar coordinate system[J]. Neurocomputing, 2017, 265: 57-65.
[2]	LEE J C, CHANG C P, CHEN W K. Detection of copy--move image forgery using histogram of orientated gradients[J]. Information Sciences, 2015, 321: 250-262.
[3]	ULIYAN D M, JALAB H A, ABDUL WAHAB A W. Image region duplication forgery detection based on angular radial partitioning and Harris key-points[J]. Symmetry, 2016, 8(7): 62.
[4]	HOSNY K M, HAMZA H M, LASHIN N A. Copy-move forgery detection of duplicated objects using accurate PCET moments and morphological operators[J]. The Imaging Science Journal, 2018, 66(6): 330-345.
[5]	SZEGEDY C, LIU W, JIA Y. Going deeper with convolutions[J/OL]. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2015, 07-12-June: 1-9. DOI:10.1109/CVPR.2015.7298594.
[6]	KUZNETSOV A, MYASNIKOV V. A new copy-move forgery detection algorithm using image preprocessing procedure[J]. Procedia engineering, 2017, 201: 436-444.
[7]	PUN C M, CHUNG J L. A two-stage localization for copy-move forgery detection[J]. Information Sciences, 2018, 463: 33-55.
[8]	JWAID M F, BARASKAR T N. Detection of Copy-Move Image Forgery Using Local Binary Pattern with Discrete Wavelet Transform and Principle Component Analysis[C/OL]//2017 International Conference on Computing, Communication, Control and Automation (ICCUBEA). 2017: 1-6. DOI:10.1109/ICCUBEA.2017.8463695.
[9]	DIXIT R, NASKAR R, SAHOO A. Copy-move forgery detection exploiting statistical image features[C/OL]//2017 International Conference on Wireless Communications, Signal Processing and Networking (WiSPNET). 2017: 2277-2281. DOI:10.1109/WiSPNET.2017.8300165.
[10]	HILAL A, HAMZEH T, CHANTAF S. Copy-move forgery detection using principal component analysis and discrete cosine transform[C/OL]//2017 Sensors Networks Smart and Emerging Technologies (SENSET). 2017: 1-4. DOI:10.1109/SENSET.2017.8125021.
[11]	SÁNCHEZ J, MONZÓN N, SALGADO DE LA NUEZ A. An analysis and implementation of the harris corner detector[J]. Image Processing On Line, 2018.
[12]	LOWE D G. Distinctive image features from scale-invariant keypoints[J]. International journal of computer vision, 2004, 60: 91-110.
[13]	MUZAFFER G, ULUTAS G. A fast and effective digital image copy move forgery detection with binarized SIFT[C/OL]//2017 40th International Conference on Telecommunications and Signal Processing (TSP). 2017: 595-598. DOI:10.1109/TSP.2017.8076056.
[14]	SHAHROUDNEJAD A, RAHMATI M. Copy-move forgery detection in digital images using affine-SIFT[C/OL]//2016 2nd International Conference of Signal Processing and Intelligent Systems (ICSPIS). 2016: 1-5. DOI:10.1109/ICSPIS.2016.7869896.
[15]	JIN G, WAN X. An improved method for SIFT-based copy–move forgery detection using non-maximum value suppression and optimized J-Linkage[J]. Signal Processing: Image Communication, 2017, 57: 113-125.
[16]	LEE J C. Copy-move image forgery detection based on Gabor magnitude[J]. Journal of visual communication and image representation, 2015, 31: 320-334.
[17]	ZANDI M, MAHMOUDI-AZNAVEH A, TALEBPOUR A. Iterative Copy-Move Forgery Detection Based on a New Interest Point Detector[J/OL]. IEEE Transactions on Information Forensics and Security, 2016, 11(11): 2499-2512. DOI:10.1109/TIFS.2016.2585118.
[18]	AMERINI I, BALLAN L, CALDELLI R. A SIFT-Based Forensic Method for Copy–Move Attack Detection and Transformation Recovery[J/OL]. IEEE Transactions on Information Forensics and Security, 2011, 6(3): 1099-1110. DOI:10.1109/TIFS.2011.2129512.
[19]	YADAV N, KAPDI R. Copy move forgery detection using SIFT and GMM[C/OL]//2015 5th Nirma University International Conference on Engineering (NUiCONE). 2015: 1-4. DOI:10.1109/NUICONE.2015.7449647.
[20]	ALBERRY H A, HEGAZY A A, SALAMA G I. A fast SIFT based method for copy move forgery detection[J]. Future Computing and Informatics Journal, 2018, 3(2): 159-165.
[21]	MOUSSA A M. A fast and accurate algorithm for copy-move forgery detection[C/OL]//2015 Tenth International Conference on Computer Engineering & Systems (ICCES). 2015: 281-285. DOI:10.1109/ICCES.2015.7393060.
[22]	LI J, LI X, YANG B. Segmentation-based image copy-move forgery detection scheme[J]. IEEE transactions on information forensics and security, 2014, 10(3): 507-518.
[23]	WANG X Y, LI S, LIU Y N. A new keypoint-based copy-move forgery detection for small smooth regions[J]. Multimedia Tools and Applications, 2017, 76: 23353-23382.
[24]	FISCHLER M A, BOLLES R C. Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography[J]. Communications of the ACM, 1981, 24(6): 381-395.
[25]	AL-HAMMADI M M, EMMANUEL S. Improving SURF Based Copy-Move Forgery Detection Using Super Resolution[C/OL]//2016 IEEE International Symposium on Multimedia (ISM). 2016: 341-344. DOI:10.1109/ISM.2016.0075.
[26]	COZZOLINO D, POGGI G, VERDOLIVA L. Efficient Dense-Field Copy–Move Forgery Detection[J/OL]. IEEE Transactions on Information Forensics and Security, 2015, 10(11): 2284-2297. DOI:10.1109/TIFS.2015.2455334.