I tried something like this, but I couldn't get any further. I want the data to be shaped in their exact columns.
table.jpg<-("C:/Users/omar/Documents/bt-etats-financiers-semestriels-30-06-2020_9.png")
data <- image_read(table.jpg) %>%
image_crop(geometry_area(0, 0, 80, 25)) %>%
image_ocr() %>%
stringi::stri_split(fixed = "\n")
I was able to extract the table from the image with the following code :
library(RDCOMClient)
library(magick)
################################################
#### Step 1 : We convert the image to a PDF ####
################################################
path_PDF <- "C:\\image_Stackoverflow78.pdf"
path_PNG <- "C:\\image_Stackoverflow78.png"
path_Word <- "C:\\image_Stackoverflow78.docx"
pdf(path_PDF, height = 12, width = 8)
im <- image_read(path_PNG)
plot(im)
abline(v = 400, col = "black")
abline(v = 2200, col = "black")
abline(v = 2400, col = "black")
abline(v = 2800, col = "black")
abline(v = 3050, col = "black")
abline(v = 3400, col = "black")
abline(h = 2000, col = "black")
abline(h = 2200, col = "black")
abline(h = 2275, col = "black")
abline(h = 2525, col = "black")
abline(h = 2600, col = "black")
abline(h = 2690, col = "black")
abline(h = 2780, col = "black")
abline(h = 2870, col = "black")
abline(h = 2960, col = "black")
abline(h = 3050, col = "black")
abline(h = 3180, col = "black")
abline(h = 3320, col = "black")
abline(h = 3850, col = "black")
abline(h = 3940, col = "black")
abline(h = 4040, col = "black")
abline(h = 4130, col = "black")
dev.off()
####################################################################
#### Step 2 : We use the OCR of Word to convert the PDF in word ####
####################################################################
wordApp <- COMCreate("Word.Application")
wordApp[["Visible"]] <- TRUE
wordApp[["DisplayAlerts"]] <- FALSE
doc <- wordApp[["Documents"]]$Open(normalizePath(path_PDF),
ConfirmConversions = FALSE)
doc$SaveAs2(path_Word)
##############################################################
#### Step 3 : We extract the table from the word document ####
##############################################################
nb_Row <- doc$tables(1)$Rows()$Count()
nb_Col <- doc$tables(1)$Columns()$Count()
mat_Temp <- matrix(NA, nrow = nb_Row, ncol = nb_Col)
for(i in 1 : nb_Row)
{
for(j in 1 : nb_Col)
{
mat_Temp[i, j] <- tryCatch(doc$tables(l)$cell(i, j)$range()$text(), error = function(e) NA)
}
}
mat_Temp
[,1] [,2] [,3]
[1,] "\r\a" "Banque de Tunisie\rEtats financiers | 30 juin 2020\rEtat de résultat | 30 Juin 2020\rEn K.TND\r\a" "\r\a"
[2,] NA NA "Notes\r\a"
[3,] "\r\a" "PRI - Intérêts et revenus assimilés\r\a" "6.1\r\a"
[4,] "\r\a" "- Commissions (en produits)\r\a" "6.2\r\a"
[5,] "\r\a" "- Gains sur portefeuille-titres commercial et opérations financières\r\a" "6.3\r\a"
[6,] "\r\a" "PR4 - Revenus du portefeuille d'investissement\r\a" "6.4\r\a"
[7,] NA "Total produits d'exploitation\r\a" "\r\a"
[8,] NA "CHI - Intérêts encourus et charges assimilées CH2 - Commissions encourues ( * )\r\a" "6.5\r6.6\r\a"
[9,] NA "Total charges d'exploitation\r\a" "\r\a"
[10,] NA "Produit net bancaire\r\a" "\r\a"
[11,] "\r\a" "PR5 CH4 - Dotations aux provisions & corrections e va eur sur creances et passifs\r\a" "6.7\r\a"
[12,] "\r\a" "PR6 CH5 - Dotations aux provisions & corrections de valeurs sur ortefeuille d'investissement\r\a" "6.8\r\a"
[13,] "\r\a" "PR 7 - Autres produits d'exploitation\r\a" "6.9\r\a"
[14,] "\r\a" "CH6 - Frais de personnel\r\a" "6.10\r\a"
[15,] "\r\a" "CH 7 - Charges générales d'exploitation (*)\r\a" "6.11\r\a"
[16,] "\r\a" "CH8 - Dotations aux amortissements sur immobilisations\r\a" "6.12\r\a"
[17,] "\r\a" "\r\a" "esu tat \r\a"
[18,] "\r\a" "PR8/CH9 - Solde en gain / perte provenant des éléments ordinaires\r\a" "6.13\r\a"
[19,] "\r\a" "CHII - Impôt sur les bénéfices\r\a" "6.14\r\a"
[20,] NA "Résultat net des activités ordinaires\r\a" "\r\a"
[21,] NA "PR9/CHIO - Solde Gain / Pertes des éléments extraordinaires\r\a" "\r\a"
[22,] "\r\a" "Résultat net de la période\r\a" "\r\a"
[23,] "\r\a" "( *) La colonne comparative de juin 2019 a été retraitée pour des fins de c \r\a" "mparab \r\a"
[24,] "\r\a" "\r\a" "\r\a"
[,4] [,5] [,6] [,7] [,8] [,9] [,10] [,11] [,12]
[1,] "\r\a" "\r\a" "\r\a" NA NA NA NA NA NA
[2,] "juin-20\r\a" "juin-19\r\a" "déc.-19\r\a" NA NA NA NA NA NA
[3,] "219 592\r\a" "227 073\r\a" "473 871\r\a" NA NA NA NA NA NA
[4,] "27 219\r\a" "29 090\r\a" "60 661\r\a" NA NA NA NA NA NA
[5,] "8 751\r\a" "19 069\r\a" "31 822\r\a" NA NA NA NA NA NA
[6,] "46 587\r\a" "31 702\r\a" "54 491\r\a" NA NA NA NA NA NA
[7,] "302 149\r\a" "306 934\r\a" "620 845\r\a" NA NA NA NA NA NA
[8,] "122 146\r1 299\r\a" "125 471\r1 372\r\a" "256 272\r3 459\r\a" NA NA NA NA NA NA
[9,] "123 445\r\a" "126 843\r\a" "259 731\r\a" NA NA NA NA NA NA
[10,] "178 704\r\a" "180 091\r\a" "361 114\r\a" NA NA NA NA NA NA
[11,] "46 390\r\a" "10 532\r\a" "36 147\r\a" NA NA NA NA NA NA
[12,] "5 825\r\a" "2 661\r\a" "12 769\r\a" NA NA NA NA NA NA
[13,] "371\r\a" "667\r\a" "1 389\r\a" NA NA NA NA NA NA
[14,] "37 366\r\a" "39 312\r\a" "74 831\r\a" NA NA NA NA NA NA
[15,] "16 049\r\a" "16 092\r\a" "32 327\r\a" NA NA NA NA NA NA
[16,] "3 889\r\a" "3 622\r\a" "7 446\r\a" NA NA NA NA NA NA
[17,] "exp 01tatlon\r\a" "\r\a" "\r\a" "\r\a" "\r\a" "\r\a" "\r\a" "\r\a" "\r\a"
[18,] "(11 197)\r\a" "(1 698)\r\a" "499\r\a" NA NA NA NA NA NA
[19,] "21 870\r\a" "31 552\r\a" "64 192\r\a" NA NA NA NA NA NA
[20,] "36 489\r\a" "75 289\r\a" "135 290\r\a" NA NA NA NA NA NA
[21,] "\r\a" "\r\a" "\r\a" NA NA NA NA NA NA
[22,] "36 489\r\a" "75 289\r\a" "135 290\r\a" NA NA NA NA NA NA
[23,] "lité (Voir note 8)\r\a" "\r\a" "\r\a" NA NA NA NA NA NA
[24,] "\r\a" "\r\a" "/ 8\r\a" NA NA NA NA NA NA