Journal of information and communication convergence engineering

The Korea Institute of Information and Commucation Engineering (한국정보통신학회)
권호 표지
DOI QR코드

DOI QR Code

Comparison of Image Duplication Detection Using the Polar Coordinates System and Histogram of Oriented Gradients Methods

  • Received : 2018.11.09
  • Accepted : 2019.01.15
  • Published : 2019.03.31
Download PDF
⟨ Previous Next ⟩

Abstract

In the current era of digital technology, and with the help of existing software, digital photo manipulation is becoming easier and faster. One example of this is the development of powerful image processing software that makes it easy for a digital image to be manipulated and edited. It is therefore very important to protect and maintain public trust in digital images. Several methods have been developed to detect image manipulation. In this paper, we compare two methods for detecting image duplication due to copy-move actions, namely the polar coordinate system and the histogram of oriented gradients methods. The former is a method based on the transfer of a Cartesian image to a polar form, making it easy to tell whether there are objects that have undergone a copy/move in an image, while the latter is a method for retrieving information related to the distribution, which uses a target in the local area as a tool to represent the shape of the target. We compare the accuracy, speed and memory usage of these two methods.

Keywords

E1ICAW_2019_v17n1_67_f0001.png 이미지

Fig. 1. Example of magnitude and theta of gradient [8].

E1ICAW_2019_v17n1_67_f0002.png 이미지

Fig. 2. PCS flowchart.

E1ICAW_2019_v17n1_67_f0003.png 이미지

Fig. 3. HOG flowchart.

E1ICAW_2019_v17n1_67_f0004.png 이미지

Fig. 4. Result of test image 1 using PCS with a block size of eight.

E1ICAW_2019_v17n1_67_f0005.png 이미지

Fig. 5. Result of test image 2 using PCS with block size of eight.

E1ICAW_2019_v17n1_67_f0006.png 이미지

Fig. 6. Result of test image 3 using PCS with block size of eight.

E1ICAW_2019_v17n1_67_f0007.png 이미지

Fig. 7. Result of test image 1 using PCS with a block size of 16.

E1ICAW_2019_v17n1_67_f0008.png 이미지

Fig. 8. Result of test image 2 using PCS with a block size of 16.

E1ICAW_2019_v17n1_67_f0009.png 이미지

Fig. 9. Result of test image 3 using PCS with a block size of 16.

E1ICAW_2019_v17n1_67_f0010.png 이미지

Fig. 10. Result of test image 1 using HOG with a block size of eight.

E1ICAW_2019_v17n1_67_f0011.png 이미지

Fig. 11. Result of test image 2 using HOG with a block size of eight.

E1ICAW_2019_v17n1_67_f0012.png 이미지

Fig. 12. Result of test image 3 using HOG with a block size of eight.

E1ICAW_2019_v17n1_67_f0013.png 이미지

Fig. 13. Result of test image 1 using HOG with a block size of 16.

E1ICAW_2019_v17n1_67_f0014.png 이미지

Fig. 14. Result of test image 2 using HOG with a block size of 16.

E1ICAW_2019_v17n1_67_f0015.png 이미지

Fig. 15. Result of test image 3 using HOG with a block size of 16.

Table 1. Accuracy of the PCS method

E1ICAW_2019_v17n1_67_t0001.png 이미지

Table 2. Processing time for the PCS method

E1ICAW_2019_v17n1_67_t0002.png 이미지

Table 3. Memory usage in the PCS method

E1ICAW_2019_v17n1_67_t0003.png 이미지

Table 4. Accuracy of the HOG method

E1ICAW_2019_v17n1_67_t0004.png 이미지

Table 5. Processing time for the HOG method

E1ICAW_2019_v17n1_67_t0005.png 이미지

Table 6. Memory usage in the HOG method

E1ICAW_2019_v17n1_67_t0006.png 이미지

References

  1. S. M. Fadl and N. A. Semary, "Robust copy-move forgery revealing in digital images using polar coordinate system," Neurocomputing, vol 265, pp. 57-65, 2017. DOI: 10.1016/j.neucom.2016.11.091.
  2. J-C. Lee, C-P. Chang, and W-K. Chen, "Detection of copy-move image forgery using histogram of orientated gradients," Information Sciences-Informatics and Computer Science, Intelligent Systems, Applications: An International Journal, vol 321, pp. 250-262, 2015. DOI: 10.1016/j.ins.2015.03.009.
  3. Video Communication Laboratory, "CoMoFoD - image database for copy-move forgery detection," [Internet], Available: http://www.vcl.fer.hr/comofod/examples.html.
  4. R. C. Gonzalez and R. E. Woods, Digital Image Processing, 4th ed, Pearson, 2017.
  5. S. Levy, "Relations between cartesian, cylindrical, and spherical coordinates," 1995, [Internet], Available: http://www.geom.uiuc.edu/docs/reference/CRC-formulas/node42.html.
  6. G. Lynch, F. Y. Shih, amd H. Y. M. Liao, "An efficient expanding block algorithm for image copy-move forgery detection," Information Sciences, vol 239, pp.253-265, 2013. DOI: 10.1016/j.ins.2013.03.028.
  7. W.Drongelen, Signal Processing for Neuroscientists, Elsevier, 2007.
  8. S. Mallick, "Histogram of oriented gradients," 2016, [Internet], Available: https://www.learnopencv.com/histogram-of-oriented-gradients/.