Skip to main content

Table 1 Selected gray-level co-occurrence matrix (GLCM) texture measures and their abbreviations and equations

From: Prediction of plant-level tomato biomass and yield using machine learning with unmanned aerial vehicle imagery

GLCM feature

Abbreviation

Formula

Sum average

SA

\(\mathop \sum \limits_{{k = 0}}^{{2(N - 1)}} k~P_{{x + y}} (k)\)

Entropy

Ent

\({-}\mathop \sum \limits_{{i = 0}}^{{N{-}1}} \mathop \sum \limits_{{j = 0}}^{{N{-}1}} P_{d} (i,~j)\log ~(P_{d} (i,~j))\)

Difference entropy

DE

\({-}\mathop \sum \limits_{{k = 0}}^{{N{-}1}} P_{{x{-}y}} (k)\log (P_{{x{-}y}} (k))\)

Sum entropy

SE

\({-}\mathop \sum \limits_{{k = 0}}^{{2\left( {N{-}1} \right)}} P_{{x + y}} (k)\log (P_{{x + y}} (k))\)

Variance

Var

\(\mathop \sum \limits_{{i = 0}}^{{N{-}1}} \mathop \sum \limits_{{j = 0}}^{{N{-}1}} (i{-}\mu )^{2} P_{d} (i,j)\)

Difference variance

DV

\(\mathop \sum \limits_{{k = 0}}^{{N{-}1}} \left( {k~{-}\mathop \sum \limits_{{k = 0}}^{{N{-}1}} k~P_{{x{-}y}} (k)} \right)^{2} P_{{x{-}y}} (k)\)

Sum variance

SV

\({-}\mathop \sum \limits_{{k = 0}}^{{2(N{-}1)}} \left( {k~{-}\mathop \sum \limits_{{k = 0}}^{{2(N{-}1)}} k~P_{{x + y}} (k)} \right)^{2} P_{{x + y}} (k)\)

Angular second moment (uniformity)

ASM

\(\mathop \sum \limits_{{i = 0}}^{{N{-}1}} \mathop \sum \limits_{{j = 0}}^{{N{-}1}} P_{d} (i,~j)^{2}\)

Inverse difference moment

IDM

\(\mathop \sum \limits_{{i = 0}}^{{N{-}1}} \mathop \sum \limits_{{j = 0}}^{{N{-}1}} \frac{1}{{1 + (i~{-}~j)^{2} }}P_{d} (i,~j)\)

Contrast

Con

\(\mathop \sum \limits_{{k = 0}}^{{N{-}1}} k^{2} P_{x-y} (k)\)

Correlation

Cor

\(\mathop \sum \limits_{{i = 0}}^{{N{-}1}} \mathop \sum \limits_{{j = 0}}^{{N{-}1}} P_{d} (i,~j)\frac{{(i~{-}~~\mu _{x} )(j~{-}~\mu _{y} )}}{{\sigma _{x} \sigma _{y} }}\)

Information measure of correlation-1

MOC-1

\(\frac{{HXY~{-}~HXY1}}{{\max (HX,~HY)}}\)

Information measure of correlation-2

MOC-2

\([1{-}\exp\{{-}2(HXY2{-}HXY)\}]^{{1/2}}\)

  1. N is the number of gray levels, \(P_d\) is the normalized symmetric GLCM dimension, \(P_{d}(i, j)\) is GLCM value on element (i, j). Other variables were calculated as shown in Additional file 1