Updating calkmeans gevi ndfiSMA tct

ytarazona · Oct 29, 2020 · 64aa89a · 64aa89a
1 parent 611bf5c
commit 64aa89a
Show file tree

Hide file tree

Showing 5 changed files with 29 additions and 20 deletions.
diff --git a/R/calkmeans.R b/R/calkmeans.R
@@ -6,7 +6,7 @@
 #' @author Yonatan Tarazona
 #'
 #' @section References:
-#' Tarazona, Y., Mar?a, Miyasiro-L?pez. (2020). Monitoring tropical forest degradation using
+#' Tarazona, Y., Maria, Miyasiro-Lopez. (2020). Monitoring tropical forest degradation using
 #' remote sensing. Challenges and opportunities in the Madre de Dios region, Peru. Remote
 #' Sensing Applications: Society and Environment, 19, 100337.
 #'
@@ -38,16 +38,19 @@
 #' data(FTdata)
 #'
 #' # Selecting the best k value
-#' best_k <- calkmeans(img = img[[1:2]], k = NULL, iter.max = 10,
+#' best_k <- calkmeans(img = image[[1:2]], k = NULL, iter.max = 10,
 #'            algo = c("Hartigan-Wong", "Lloyd", "Forgy", "MacQueen"), iter = 30)
 #' # Jambu Elbow
 #' plot(best_k)
 #'
 #'
 #'# Selecting the best embedded algorithm in kmeans
-#' best_algo <- calkmeans(img = img[[1:2]], k = 4, iter.max = 10,
+#' best_algo <- calkmeans(img = image[[1:2]], k = 4, iter.max = 10,
 #'            algo = c("Hartigan-Wong", "Lloyd", "Forgy", "MacQueen"), iter = 30)
 #'
+#' # Choose the algorithm with the highest value
+#' best_algo
+#'
 #' @export
 #'
 
@@ -191,7 +194,7 @@ calkmeans <- function(img, k = NULL, iter.max = 10, algo = c("Hartigan-Wong", "L
 
     resulFinal <- c(vecIner.hw, vecIner.l, vecIner.f, vecIner.m)
 
-    return(structure(resulFinal, class = "calkmeans"))
+    return(resulFinal)
 
   }
 
@@ -202,6 +205,7 @@ calkmeans <- function(img, k = NULL, iter.max = 10, algo = c("Hartigan-Wong", "L
 
 #' Plot for the "calkmeans" class
 #'
+#' @export
 #'
 plot.calkmeans <- function(x, xlab, ylab, type, main, cex, ...){
 

diff --git a/R/gevi.R b/R/gevi.R
@@ -19,9 +19,11 @@
 #' Sentinel2MSI \tab coastal,blue,green,red,nir-1,mir-1,mir-2 \cr
 #' }
 #'
-#' @param x It could be RasterStack or RasterBrick.
+#' @param img It could be RasterStack or RasterBrick.
 #' @param sat Specify satellite and sensor type (Landsat5TM, Landsat7ETM or Landsat8OLI).
 #'
+#' @importFrom raster as.matrix
+#'
 #' @examples
 #' library(ForesToolboxRS)
 #' library(raster)
@@ -30,11 +32,16 @@
 #' data(FTdata)
 #'
 #' # Tasseled-cap using Landsat8OLI
-#' tasscap <- tct(x = img, sat = "Landsat8OLI")
+#' gevi_index <- gevi(img = image/10000, sat = "Landsat8OLI")
+#'
+#' # Improving the plot histogram
+#' gevi_index[gevi_index > 1] <- NA
+#' gevi_index[gevi_index < -1] <- NA
+#' plot(gevi_index)
 #'
 #' @export
 #'
-gevi <- function(x, sat = "Landsat8OLI"){
+gevi <- function(img, sat = "Landsat8OLI"){
 
   if (sat == "Landsat4TM"){
     coefc <- matrix(c(0.3037,  0.2793,  0.4743, 0.5585, 0.5082,   0.1863,
@@ -74,7 +81,7 @@ gevi <- function(x, sat = "Landsat8OLI"){
                                                  c("B1","B2","B3","B4","B8","B11","B12")))
   } else stop("Satellite not supported.", call. = TRUE)
 
-  val <- as.matrix(x)%*%t(coefc)
+  val <- as.matrix(img)%*%t(coefc)
 
   bgw <- img[[1:3]]
 

diff --git a/R/ndfiSMA.R b/R/ndfiSMA.R
@@ -35,22 +35,21 @@
 #' data(FTdata)
 #'
 #' # Unmix the image
-#' ndfi <- ndfiSMA(img, procesLevel="SR")
+#' ndfi <- ndfiSMA(img = image, procesLevel="SR")
 #' plot(ndfi)
 #'
 ndfiSMA <- function(img, procesLevel="SR", verbose = FALSE){
 
   if (is(img, "RasterStack") | is(img, "RasterBrick")) {
     df <- as.matrix(img)
   } else {
-    stop(class(img), " This class is not supported yet.", call. = TRUE)
+    stop(class(img), ": This class is not supported yet. It must be RasterStack or RasterBrick", call. = TRUE)
   }
 
   if (procesLevel=="SR") {
 
     if(verbose){
       message(paste0(paste0(rep("*",10), collapse = ""), " Endmembers in surface reflectance " , paste0(rep("*",10), collapse = "")))
-      print(model_algo)
     }
 
     # Endmembers
@@ -66,7 +65,6 @@ ndfiSMA <- function(img, procesLevel="SR", verbose = FALSE){
 
     if(verbose){
       message(paste0(paste0(rep("*",10), collapse = ""), " Endmembers in TOA value " , paste0(rep("*",10), collapse = "")))
-      print(model_algo)
     }
 
     # Endmembers
@@ -85,7 +83,6 @@ ndfiSMA <- function(img, procesLevel="SR", verbose = FALSE){
 
   if(verbose){
     message(paste0(paste0(rep("*",10), collapse = ""), " Obtaining fractions through least squares " , paste0(rep("*",10), collapse = "")))
-    print(model_algo)
   }
 
   # Spectral Mixture Analysis
@@ -114,7 +111,6 @@ ndfiSMA <- function(img, procesLevel="SR", verbose = FALSE){
 
   if(verbose){
     message(paste0(paste0(rep("*",10), collapse = ""), " Obtaining the NDFI index " , paste0(rep("*",10), collapse = "")))
-    print(model_algo)
   }
 
   gv <- fractions[,1]*100; gv[gv < 0] <- 0 # Green Vegetation

diff --git a/R/sma.R b/R/sma.R
@@ -63,7 +63,6 @@ sma <- function(img, endm, verbose = FALSE){
 
   if(verbose){
     message(paste0(paste0(rep("*",10), collapse = ""), " Obtaining fractions through least squares " , paste0(rep("*",10), collapse = "")))
-    print(model_algo)
   }
 
   if(dim(df)[2] > dim(endm)[1]){
@@ -79,7 +78,6 @@ sma <- function(img, endm, verbose = FALSE){
 
       if(verbose){
         message(paste0(paste0(rep("*",10), collapse = ""), " Obtaining Root Mean Square Error " , paste0(rep("*",10), collapse = "")))
-        print(model_algo)
       }
 
       # We estimate the RMSE
@@ -89,7 +87,6 @@ sma <- function(img, endm, verbose = FALSE){
 
       if(verbose){
         message(paste0(paste0(rep("*",10), collapse = ""), " Save the fractions on a raster " , paste0(rep("*",10), collapse = "")))
-        print(model_algo)
       }
 
       # We store the fractions on a raster

diff --git a/R/tct.R b/R/tct.R
@@ -38,6 +38,8 @@
 #' @param sat Specify satellite and sensor type (Landsat5TM, Landsat7ETM
 #' or Landsat8OLI).
 #'
+#' @importFrom raster as.matrix
+#'
 #' @examples
 #' library(ForesToolboxRS)
 #' library(raster)
@@ -46,11 +48,14 @@
 #' data(FTdata)
 #'
 #' # Tasseled-cap using Landsat8OLI
-#' sat_tct <- tct(img, sat="Landsat8OLI")
+#' sat_tct <- tct(img = image, sat="Landsat8OLI")
+#' plotRGB(sat_tct, 1,2,3, stretch="lin")
 #'
 #'@export
 #'
-tct <- function(x=img, sat="Landsat8OLI"){
+tct <- function(img, sat="Landsat8OLI"){
+
+  if(!inherits(img, "Raster")) stop("img must be a RasterBrick or RasterStack", call. = TRUE)
 
   if (sat=="Landsat4TM"){
     coefc <- matrix(c(0.3037,  0.2793,  0.4743, 0.5585, 0.5082,   0.1863,
@@ -90,7 +95,7 @@ tct <- function(x=img, sat="Landsat8OLI"){
                                                 c("B1","B2","B3","B4","B8","B11","B12")))
   } else stop("Satellite not supported.")
 
-  val <- as.matrix(x)%*%t(coefc)
+  val <- as.matrix(img)%*%t(coefc)
 
   bgw <- img[[1:3]]