Table 9.

Introduction of image segmentation performance evaluation parameters.

Indicators	Formulation	Remark
PSNR	$PSNR=20·{\text{log}}_{10}\left[\frac{255}{\sqrt{\frac{{\sum }_{i=0}^{M-1}{\sum }_{j=0}^{N-1}{\left(I_{\mathit{ij}},-,S,e,g_{\mathit{ij}}\right)}^2}{M\times N}}}\right]$	Compare the split image to the original image and evaluate the differences
SSIM	$\text{SSIM}=\frac{\left(2,{\mu }_I,{\mu }_{\text{Seg}},+,c_1\right)\left(2,{\sigma }_{I,\text{Seg}},+,c_2\right)}{\left({\mu }_I^2,+,{\mu }_{\mathit{Seg}}^2,+,c_1\right)\left({\sigma }_I^2,+,{\sigma }_{\mathit{Seg}}^2,+,c_2\right)}$	A comparison with the segmented, non-distorted, uncompressed image is performed to determine differences and similarities
FSIM	$FSIM=\frac{{\sum }_{I\in \mathrm{\Omega }}{S}_L(X)P{C}_m(X)}{{\sum }_{I\in \mathrm{\Omega }}P{C}_m(X)}$	Outlining the quality score which demonstrates the importance of a local structure