RGB Visible Indices for Image Analysis

Corresponding author: santoshagriculture@gmail.com

1. Introduction

The rgbIndices package provides a comprehensive set of RGB-based indices derived from digital images. These indices are widely used in agriculture, crop phenotyping, vegetation monitoring, and image-based modeling.

The package includes multiple groups of indices such as basic, difference, ratio, normalized difference, vegetation, and color indices.

---

2. RGB Components

An RGB image consists of three channels:

• R: Red (0–255)
  
• G: Green (0–255)
  
• B: Blue (0–255)

---

3. Basic Indices

Index	Formula
Normalized Red (r)	R / (R + G + B)
Normalized Green (g)	G / (R + G + B)
Normalized Blue (b)	B / (R + G + B)
Intensity (INT)	(R + G + B) / 3

---

4. Difference Indices

Index	Full Form	Formula
GRD	Green Red Difference	G − R
BGD	Blue Green Difference	B − G
RGD	Red Green Difference	R − G
RBD	Red Blue Difference	R − B
GBD	Green Blue Difference	G − B
BRD	Blue Red Difference	B − R

---

5. Ratio Indices

Index	Full Form	Formula
GRRI	Green Red Ratio Index	G / R
GBRI	Green Blue Ratio Index	G / B
RBRI	Red Blue Ratio Index	R / B
RGRI	Red Green Ratio Index	R / G
BGRI	Blue Green Ratio Index	B / G
BRRI	Blue Red Ratio Index	B / R

---

6. Normalized Difference Indices

Index	Full Form	Formula
NGRDI	Normalized Green Red Difference Index	(G − R) / (R + G + B)
NRGDI	Normalized Red Green Difference Index	(R − G) / (R + G + B)
NBRDI	Normalized Blue Red Difference Index	(B − R) / (R + G + B)
NRBDI	Normalized Red Blue Difference Index	(R − B) / (R + G + B)
NGBDI	Normalized Green Blue Difference Index	(G − B) / (R + G + B)
NBGDI	Normalized Blue Green Difference Index	(B − G) / (R + G + B)

---

Note: Some normalized difference indices are sign-inverted counterparts of each other (e.g., NGRDI vs NRGDI).

7. Vegetation Indices

Index	Full Form	Formula
WI	Woebbecke Index	(G − B) / (R − G)
GRVI	Green Red Vegetation Index	(G − R) / (G + R)
IKAW	Kawashima Index	(R − B) / (R + B)
NDTI	Normalized Difference Turbidity Index	(R − G) / (R + G)
GBI	Green Blue Index	(G − B) / (G + B)
GLI	Green Leaf Index	(2G − R − B) / (2G + R + B)
VARI	Visible Atmospherically Resistant Index	(G − R) / (G + R − B)
NDI	Normalized Difference Index	(g − r) / (g + r)
ExG	Excess Green Index	2g − r − b
ExR	Excess Red Index	1.4r − g
ExGR	Excess Green minus Red	3g − 2.4r − b
MxEG	Modified Excess Green	1.262G − 0.884R − 0.311B
ExB	Excess Blue	1.4b − g
RGBVI	RGB Vegetation Index	(G² − RB) / (G² + RB)

---

8. Color Indices

Index	Full Form	Formula
Grey	Gray Intensity	0.2898r + 0.5870g + 0.1140b
BI	Brightness Index	√((R² + G² + B²)/3)
HI	Hue Index	(2R − G − B) / (G − B)
RI	Redness Index	R² / (B × G³)
SI	Saturation Index	2(R − G − B) / (G − B)
CI	Coloration Index	(R − B) / R
CIVE	Color Index of Vegetation	0.441R − 0.811G + 0.385B + 18.78745
VEG	Vegetative Index	G / (R^0.667 × B^0.333)
SAT	Overall Saturation Index	(
OHI	Overall Hue Index	atan(2(R − G − B)/(30.5(G − B)))
TCVI	True Color Vegetation Index	1.4(2R − 2B)/(2R − G − 2B + 255×0.4)

---

9. Example Workflow

library(rgbIndices)
library(raster)

## Loading required package: sp

# ---------------------------
# Example
# ---------------------------
set.seed(123)
r <- raster::raster(matrix(runif(30*30), 30, 30))
g <- raster::raster(matrix(runif(30*30), 30, 30))
b <- raster::raster(matrix(runif(30*30), 30, 30))
img <- raster::stack(r, g, b)

# Compute indices
idx  <- rgb_basic(img)
idx1 <- rgb_diff(img)
idx2 <- rgb_ratio(img)
idx3 <- rgb_normdiff(img)
idx4 <- rgb_veg(img)
idx5 <- rgb_color(img)

# Summary statistics
rgb_indices_to_mean(idx)

## # A tibble: 1 × 7
##       R     G     B     r     g     b   INT
##   <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 0.496 0.500 0.493 0.334 0.336 0.330 0.497

# Convert to table
tbl <- rgb_indices_to_tbl(idx)
head(tbl)

## # A tibble: 6 × 7
##       R      G     B     r     g      b   INT
##   <dbl>  <dbl> <dbl> <dbl> <dbl>  <dbl> <dbl>
## 1 0.288 0.924  0.155 0.210 0.676 0.114  0.456
## 2 0.963 0.887  0.864 0.355 0.327 0.318  0.904
## 3 0.665 0.508  0.575 0.380 0.291 0.329  0.583
## 4 0.131 0.0872 0.541 0.172 0.115 0.713  0.253
## 5 0.648 0.461  0.784 0.342 0.244 0.414  0.631
## 6 0.847 0.709  0.136 0.501 0.419 0.0804 0.564

—————————

Real image example

—————————

img_real <- raster::stack(rgb_example())
raster::plotRGB(img_real)
rgb_basic(img_real)

---

10. Applications

• Crop health monitoring
  
• Disease detection
  
• Plant phenotyping
  
• Precision agriculture
  
• Weed detection and classification

11. Reference

Singh, R. N., Krishnan, P., Singh, V. K., & Das, B. (2023).
Estimation of yellow rust severity in wheat using visible and thermal imaging coupled with machine learning models.
Geocarto International.
https://www.tandfonline.com/doi/full/10.1080/10106049.2022.2160831