Image recognition in R using convolutional neural networks with the MXNet package. Part 2. Full article at: https://firsttimeprogrammer.blogspot.com/2016/07/image-recognition-in-r-using.html

im_rec_2.R

# Generate a train-test dataset

# Clean environment and load required packages
rm(list=ls())
require(EBImage)

# Set wd where resized greyscale images are located
setwd("C://dogs_resized")

# Out file
out_file <- "C://dogs_28.csv"

# List images in path
images <- list.files()

# Set up df
df <- data.frame()

# Set image size. In this case 28x28
img_size <- 28*28

# Set label
label <- 1

# Main loop. Loop over each image
for(i in 1:length(images))
{
    # Read image
    img <- readImage(images[i])
    # Get the image as a matrix
    img_matrix <- [email protected]
    # Coerce to a vector
    img_vector <- as.vector(t(img_matrix))
    # Add label
    vec <- c(label, img_vector)
    # Bind rows
    df <- rbind(df,vec)
    # Print status info
    print(paste("Done ", i, sep = ""))
}

# Set names
names(df) <- c("label", paste("pixel", c(1:img_size)))

# Write out dataset
write.csv(df, out_file, row.names = FALSE)

#-------------------------------------------------------------------------------
# Test and train split and shuffle

# Load datasets
plants <- read.csv("plants_28.csv")
dogs <- read.csv("dogs_28.csv")

# Bind rows in a single dataset
new <- rbind(plants, dogs)

# Shuffle new dataset
shuffled <- new[sample(1:1512),]

# Train-test split
train_28 <- shuffled[1:1200,]
test_28 <- shuffled[1201:1512,]

# Save train-test datasets
write.csv(train_28, "train_28.csv",row.names = FALSE)
write.csv(test_28, "test_28.csv",row.names = FALSE)

Nice example and article on your blog. How could we download the 1500 images of dogs and plants?

Hi! Thank you for the comment! I usually link to the datasets that I use but this time since I do not own any of the images I couldn't do it. You can get the links to each image at image-net.org, they are mostly on Tumblr (publicly available) or similar websites. I used about 1500 of those at the following links http://www.image-net.org/synset?wnid=n02084071 and http://www.image-net.org/synset?wnid=n00017222 however not all are .jpeg so if you want to use all of them you might need to do some more preprocessing.

Thanks for the great code. I was just wondering whether you can give an example of the columns available in the "plants_28.csv" and "dogs_28.csv" files. I think it is not immediate for me to understand how to send the classes info to the ANN.

Hi Sciabola, I've just released a step by step tutorial on how to set up and run a deep convolutional neural network with a publicly available dataset. If you like, you can check it at https://firsttimeprogrammer.blogspot.com/2016/08/image-recognition-tutorial-in-r-using.html