Czenakowski distance

The Czenakowski distance (sometimes shortened as CZD) is a per-pixel quality metric that estimates quality or similarity by measuring differences between pixels. Because it compares vectors with strictly non-negative elements, it is often used to compare colored images, as color values cannot be negative. This different approach has a better correlation with subjective quality assessment than PSNR.^{[ citation needed ]}

Definition

Androutsos et al. give the Czenakowski coefficient as follows: ^[1]

${\displaystyle d_{z}(i,j)=1-{\frac {2\sum _{k=1}^{p}{\text{min}}(x_{ik},\ x_{jk})}{\sum _{k=1}^{p}(x_{ik}+x_{jk})}}}$

Where a pixel ${\displaystyle x_{i}}$ is being compared to a pixel ${\displaystyle x_{j}}$ on the k-th band of color – usually one for each of red, green and blue.

For a pixel matrix of size ${\displaystyle M\times N}$, the Czenakowski coefficient can be used in an arithmetic mean spanning all pixels to calculate the Czenakowski distance as follows: ^{[2] [3]}

${\displaystyle {\frac {1}{MN}}\sum _{i=0}^{M-1}\sum _{j=0}^{N-1}{\begin{pmatrix}1-{\frac {2\sum _{k=1}^{3}{\text{min}}(A_{k}(i,j),\ B_{k}(i,j))}{\sum _{k=1}^{3}(A_{k}(i,j)+B_{k}(i,j))}}\end{pmatrix}}}$

Where ${\displaystyle A_{k}(i,j)}$ is the (i, j)-th pixel of the k-th band of a color image and, similarly, ${\displaystyle B_{k}(i,j)}$ is the pixel that it is being compared to.

Uses

In the context of image forensics – for example, detecting if an image has been manipulated –, Rocha et al. report the Czenakowski distance is a popular choice for Color Filter Array (CFA) identification. ^[2]

References

1. Androutsos, D.; Plataniotiss, K.N.; Venetsanopoulos, A.N. (1998). "Distance measures for color image retrieval". Proceedings 1998 International Conference on Image Processing. ICIP98 (Cat. No.98CB36269). Vol. 2. pp. 770–774. doi:10.1109/ICIP.1998.723652. ISBN   0-8186-8821-1. S2CID   10134889.
2. Rocha, Anderson; Scheirer, Walter; Boult, Terrance; Goldenstein, Siome (October 2011). "Vision of the unseen" . ACM Computing Surveys. 43 (4): 1–42. doi:10.1145/1978802.1978805. ISSN   0360-0300. S2CID   113533.
3. Kale, K. V.; Mehrotra, S. C.; R. R. Manza, R. R., eds. (2007). Advances in Computer Vision and Information Technology. New Delhi, India: I.K. International Pvt. Ltd. p. 91. ISBN   978-81-89866-74-7.