Table 1.

Features extracted from RGB and hyperspectral images for disease diagnosis.

Feature	Description	Equation	Reference
Descriptive color statistical features (1–36)
Mean	he average value of the data.	$\bar{\mathrm{x}}=\frac{1}{\mathrm{n}}\sum _{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{x}}_{\mathrm{i}}$	Ashfaq et al. [24]
Maximum	The largest value in the data.	$M=\max \left(X\right)iff\left\{a\in X\cap \forall x\in X,x\le a\right\}$	Ashfaq et al. [24]
Standard Deviation	A measure of the spread of the data.	$\mathrm{s}=\sqrt{\frac{1}{\mathrm{n}-1}\sum _{\mathrm{i}=1}^{\mathrm{n}}{\left(\mathrm{x}-\bar{\mathrm{x}}\right)}^2}$	Ashfaq et al. [24]
Median	The number in the middle of the data set when it is sorted in increasing order.	$Med\left(X\right)=\{\begin{array}{l}X\left[\frac{n+1}{2}\right]ifnisodd\\ \frac{X\left[\frac{n}{2}\right]+X\left[\frac{n}{2}+1\right]}{2}ifniseven\end{array}$	Ashfaq et al. [24]
Textural features (37–154)
Contrast	This is a measure of the intensity difference between neighboring pixels.	${\sum _{\mathrm{i},\mathrm{j}}\left|\mathrm{i}-\mathrm{j}\right|}^2\mathrm{p}\left(\mathrm{j},\mathrm{j}\right)$	Haralick et al. [25]
Correlation	The degree of linear dependence between neighboring pixels.	$\sum _{\mathrm{i},\mathrm{j}}\frac{\left(\mathrm{i}-\mu \mathrm{i}\right)\left(\mathrm{j}-\mu \mathrm{j}\right)\mathrm{p}\left(\mathrm{i},\mathrm{j}\right)}{{\mathrm{\sigma }}_{\mathrm{i}}{\mathrm{\sigma }}_{\mathrm{j}}}$	Haralick et al. [25]
Energy	The sum of the squares of all the possible patterns of intensity difference between neighboring pixels in all directions.	$\sum _{\mathrm{i},\mathrm{j}}\mathrm{p}{\left(\mathrm{i},\mathrm{j}\right)}^2$	Haralick et al. [25]
Homogeneity	Measures the uniformity of the neighboring pixels	$\sum _{\mathrm{i},\mathrm{j}}\frac{\mathrm{p}\left(\mathrm{i},\mathrm{j}\right)}{1+\left|\mathrm{i}-\mathrm{j}\right|}$	Haralick et al. [25]
Mean	The average intensity value of neighboring pixels.	$\frac{1}{\mathrm{n}}\left({\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{X}}_{\mathrm{i}}\right)$	Haralick et al. [25]
Standard Deviation	Measures the spread of the intensity values of the neighboring pixels.	$\sqrt{\frac{1}{\mathrm{n}}\left(\sum _{\mathrm{i}=1}^{\mathrm{n}}{\left({\mathrm{X}}_{\mathrm{i}}-\bar{\mathrm{X}}\right)}^2\right)}$	Haralick et al. [25]
Entropy	Measure of the amount of information stored in the texture image.	$\mathrm{E}=\text{sum}\left(\mathrm{p}.\ast \log 2\left(\mathrm{p}\right)\right)$	Haralick et al. [25]
RMS	Measures the square root of the average of the sum of the squares of the difference between intensity values of neighboring pixels.	${\mathrm{X}}_{\text{RMS}}=\sqrt{\frac{1}{\mathrm{N}}{\sum _{\mathrm{n}=1}^{\mathrm{N}}\left|{\mathrm{X}}_{\mathrm{n}}\right|}^2}$	Haralick et al. [25]
Variance	A measure of the spread of the intensity values of the neighboring pixels.	$\frac{1}{\mathrm{n}}\left({\sum _{\mathrm{i}=1}^{\mathrm{n}}\left({\mathrm{X}}_{\mathrm{i}}-\bar{\mathrm{X}}\right)}^2\right)$	Haralick et al. [25]
Smoothness	Measures the degree to which the neighboring pixels have similar intensity values.	–	Haralick et al. [25]
Kurtosis	A measure of the peakedness of the distribution of the intensity values of the neighboring pixels.	$\mathrm{K}=\frac{\mathrm{E}{\left(\mathrm{x}-\mathrm{\mu }\right)}^4}{{\mathrm{\sigma }}^4}$	Haralick et al. [25]
Skewness	A measure of the asymmetric distribution of the intensity values of the neighboring pixels.	$\mathrm{S}=\frac{\mathrm{E}{\left(\mathrm{x}-\mathrm{\mu }\right)}^3}{{\mathrm{\sigma }}^3}$	Haralick et al. [25]
IDM	A sophisticated GLCM feature that captures complex texture features in the neighborhood of a pixel.	–	Haralick et al. [25]
Shape features (155–166)
Area	The mean of number of pixels in each region segmented in the image	$\mathrm{a}=\sum _{\mathrm{u}=1}^{\mathrm{M}}\sum _{\mathrm{v}=1}^{\mathrm{N}}\mathrm{A}\left[\mathrm{u},\mathrm{v}\right]$	Vishnoi et al. (2021)
Perimeter	The mean of number of boundary pixels on each region segmented in the image	2 (length + width)	Vishnoi et al. (2021)
Number of objects	Number of disease spots in the leaf image	–	–
Major axis length	The distance along the longest axis of an object, in pixels or other appropriate units.	$\mathrm{M}={\mathrm{x}}_1+{\mathrm{x}}_2$	Vishnoi et al. (2021)
Minor axis length	The distance along the shortest axis of an object, in pixels or other appropriate units.	$\mathrm{m}=\sqrt{\left({\left({\mathrm{x}}_1+{\mathrm{x}}_2\right)}^2-\mathrm{d}\right)}$	Vishnoi et al. (2021)
eccentricity index	A measurement of the difference between the lengths of the major and minor axis, normalized to range between 0 and 1.	$\frac{\text{Major}\text{axis}\text{length}}{\text{Minor}\text{axis}\text{length}}$	Vishnoi et al. (2021)
Circularity	A ratio of a circle to the bounding box of an object, with a value of 1 indicating a perfect circle and values approaching 0 indicating a more complex or amorphous shape.	$\frac{4\times \text{pi}\times \text{area}}{{\begin{array}{l}\text{Perimeter}\end{array}}^2}$	Vishnoi et al. (2021)
Circular index	Mean of circularity of each segmented region in the image	$\frac{{\text{Perimeter}}^2}{\text{Area}}$	Javidan et al. [26]
Compactness index	Mean of circularity of each segmented region in the image	$\frac{\text{Area}}{{\text{Perimeter}}^2}$	Javidan et al. [26]
Shape factors 1	Dimensionless numerical values calculated based on at least two simple shape measures, which makes them independent to object orientation, translation, and scale. These features may provide useful information for the morphological description of plants. Primary shape features, namely area, perimeter, and major and minor axis length values of each plant are extracted from binary images	$\frac{\begin{array}{l}\text{Major}\text{axis}\text{length}\end{array}}{\begin{array}{l}\text{Area}\end{array}}$	Javidan et al. [26]
Shape factors 2		$\frac{\begin{array}{l}\text{Area}\end{array}}{\begin{array}{l}\text{Major}\text{axis}{\text{length}}^3\end{array}}$	Javidan et al. [26]
Shape factors 3		$\frac{\begin{array}{l}4\times \text{Area}\end{array}}{\begin{array}{l}\text{pi}\times \text{Major}\text{axis}\text{length}\times \text{Major}\text{axis}\text{length}\end{array}}$	Javidan et al. (2024)