Table 1 Summary of Haralick features
Feature name | Feature description | Feature formula |
|---|---|---|
Angular second moment (ASM) | Shows how uniform a texture is by measuring local homogeneity | \(ASM=\sum _{i}\sum _{j}p\left (i,j\right)^{2}\) |
Energy (E) | Measures homogeneity | \( E=\sqrt {\sum _{i}\sum _{j}p\left (i,j\right)^{2}}\) |
Contrast (C) | Shows variation in texture | \(C=\sum _{i}\sum _{j}|i-j|^{2}p\left (i,j\right)\) |
Dissimilarity (D) | Variation in texture | \(D=\sum _{i}\sum _{j}|i-j|p\left (i,j\right)\) |
Homogeneity (H) | Uniformity of non-zero entries | \(H=\sum _{i}\sum _{j}\frac {1}{1+\left (1-j\right)^{2}}p\left (i,j\right)\) |
Entropy (En) | Spatial disorder of texture | \(En=\sum _{i}\sum _{j}p\left (i,j\right)\log \left (p\left (i,j\right)\right)\) |
Correlation (Cr) | Linear relationship of texture | \(Cr=\sum _{i}\sum _{j}p\left (i,j\right)\frac {\left (i-\mu _{x}\right)\left (j-\mu _{y}\right)}{\sigma _{x} \sigma _{y}}\) |
Autocorrelation (ACr) | Measure repeating patterns | \(ACr=\sum _{i}\sum _{j}\left (i \cdot j\right)p\left (i,j\right)\) |
Variance (V) | Measure of texture heterogeneity | \(V=\sum _{i}\sum _{j}\left (i-\mu _{x}\right)^{2}\cdot p\left (i,j\right)+\sum _{i}\sum _{j}\left (i-\mu _{y}\right)^{2}\cdot p\left (i,j\right)\) |
Cluster shade (Cs) | Measure perceptual uniformity | \(Cs=\sum _{i}\sum _{j}\left (i+j-\mu _{x}-\mu _{y}\right)^{3}p\left (i,j\right)\) |
Cluster prominence (Cp) | Measure image symmetry | \(Cp=\sum _{i}\sum _{j}\left (i+j-\mu _{x}-\mu _{y}\right)^{4}p\left (i,j\right)\) |
Maximum probability (Mp) | Maximum co-occurrence | Mp= maxp(i,j) |