Table 19 List of histogram features
From: Feature analysis for classification of trace fluorescent labeled protein crystallization images
Symbol | Description | Formulation |
|---|---|---|
μ | Average image intensity | \(\frac {1}{w*h}{\sum \nolimits }_{k=0}^{k=255}{k * H[k]}\) |
σ | Std devn of intensity | \(\sqrt {\frac {1}{w*h}{\sum \nolimits }_{k=0}^{k=255}{ (k - \mu)^ 2 * H[k] }}\) |
s | Skewness | \(\frac {1}{(w*h)*\sigma ^{1.0}} {\sum \nolimits }_{k=0}^{k=255}{ (k - \mu)^ 3 * H[k]}\) |
k | Kurtosis | \(\frac {1}{(w*h)*\sigma ^{2}} {\sum \nolimits }_{k=0}^{k=255}{ (k - \mu)^ 4 * H[k] }\) |
vE | Entropy | -\({\sum \nolimits }_{k=0}^{255} N[k] log(N[k]) \text {, where} N[k] = H[k]/(w*h)\) |
\(g_{1}^{1}\), \(g_{1}^{2}\), \(g_{1}^{3}\),.. \(g_{3}^{3}\) | GLCM auto-correlation | Eq. 5 |
i a 1, i a 2, i a 3 | Image auto-correlation | Eq. 6 |
m g 1, m g 2, m g 3 | GLCM power spectrum magnitude | m g i =m e a n2(|f f t s h i f t(f f t2(P i ))|),1≤i≤3 |
mi | Image power spectrum magnitude | m i=m e a n2(|f f t s h i f t(f f t2(I))|.2) |