Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

An Effective Image Copy-Move Forgery Detection Using Entropy Information

Li Jiang, Zhaowei Lu

TL;DR

This work tackles copy-move forgery detection by addressing the limitations of traditional keypoint-based CMFD in smooth regions. It introduces entropy images to guide the SIFT detector, producing more informative keypoints and scales, and proposes an overlapped entropy level clustering scheme to curb the combinatorial cost of matching. The method demonstrates strong detection performance across public high- and mid-resolution datasets, with a careful parameter study showing that entropy-based pre-processing and clustering yield superior accuracy while maintaining practical runtimes. Overall, the entropy-informed framework advances CMFD robustness and efficiency, particularly for high-resolution images with low-texture tampered regions.

Abstract

Image forensics has become increasingly crucial in our daily lives. Among various types of forgeries, copy-move forgery detection has received considerable attention within the academic community. Keypoint-based algorithms, particularly those based on Scale Invariant Feature Transform, have achieved promising outcomes. However, most of keypoint detection algorithms failed to generate sufficient matches when tampered patches were occurred in smooth areas, leading to insufficient matches. Therefore, this paper introduces entropy images to determine the coordinates and scales of keypoints based on Scale Invariant Feature Transform detector, which make the pre-processing more suitable for solving the above problems. Furthermore, an overlapped entropy level clustering algorithm is developed to mitigate the increased matching complexity caused by the non-ideal distribution of gray values in keypoints. Experimental results demonstrate that our algorithm achieves a good balance between performance and time efficiency.

An Effective Image Copy-Move Forgery Detection Using Entropy Information

TL;DR

This work tackles copy-move forgery detection by addressing the limitations of traditional keypoint-based CMFD in smooth regions. It introduces entropy images to guide the SIFT detector, producing more informative keypoints and scales, and proposes an overlapped entropy level clustering scheme to curb the combinatorial cost of matching. The method demonstrates strong detection performance across public high- and mid-resolution datasets, with a careful parameter study showing that entropy-based pre-processing and clustering yield superior accuracy while maintaining practical runtimes. Overall, the entropy-informed framework advances CMFD robustness and efficiency, particularly for high-resolution images with low-texture tampered regions.

Abstract

Image forensics has become increasingly crucial in our daily lives. Among various types of forgeries, copy-move forgery detection has received considerable attention within the academic community. Keypoint-based algorithms, particularly those based on Scale Invariant Feature Transform, have achieved promising outcomes. However, most of keypoint detection algorithms failed to generate sufficient matches when tampered patches were occurred in smooth areas, leading to insufficient matches. Therefore, this paper introduces entropy images to determine the coordinates and scales of keypoints based on Scale Invariant Feature Transform detector, which make the pre-processing more suitable for solving the above problems. Furthermore, an overlapped entropy level clustering algorithm is developed to mitigate the increased matching complexity caused by the non-ideal distribution of gray values in keypoints. Experimental results demonstrate that our algorithm achieves a good balance between performance and time efficiency.
Paper Structure (13 sections, 18 equations, 7 figures, 3 tables)

This paper contains 13 sections, 18 equations, 7 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Related technique
  3. Review Of The SIFT
  4. Detect Smooth Or Small Tampering
  5. Proposed method
  6. Pre-processing
  7. Hierarchical Keypoint Clustering Matching
  8. Experiments
  9. Datasets
  10. Evaluation Metrics
  11. Analysis Of Parameters
  12. Performance Comparison On Different Datasets
  13. Conclusion

Figures (7)

  • Figure 1: Keypoint detection by (a) classical contrast threshold; (b) reducing contrast threshold.
  • Figure 2: Pre-processing results of different types of images. (a) RGB image; (b) the gradient information of gray image; (c) the gradient information of entropy image; (d) the number of keypoints within $16 \times 16$ in gray images; (e) the number of keypoints within $16 \times 16$ in entropy images.
  • Figure 3: Framework of the proposed algorithm.
  • Figure 4: An example of hierarchical feature point clustering. (a) the gray value distribution of the keypoints; (b) the entropy value distribution of the keypoints.
  • Figure 5: Some experimental results. (a) forgery images; (b) matching; (c) binary result; (d) ground-true.
  • ...and 2 more figures