Loading a directory of images to use

This is something that is very useful all the time when using computer vision. You may not have a camera readily available but you can easily load a directory of images to play with.

Sometimes sets of images are nice to have as well. You can find data sets of similiar things, for instance maybe a bunch of pictures of fruit and you want to use computer vision to detect the type of fruit based on features that we define.

Note

Free data sets are available in a list here: http://github.com/sightmachine/SimpleCV/wiki/List-of-know-open-data-sets-for-testing

So what you will need to do is browse the web and download images into a seperate directory or in the same directory as the script. For this example the file extensions have to be .png, but you can change the code to .jpg, .bmp, etc.

To load up and show a directory of images lets look at the code:

import os
import glob
import time
from SimpleCV import *

print __doc__

#Settings
my_images_path = "/tmp/cats/" #put your image path here if you want to override current directory
extension = "*.png"


#Program
if not my_images_path:
        path = os.getcwd() #get the current directory
else:
        path = my_images_path

imgs = list() #load up an image list
directory = os.path.join(path, extension)
files = glob.glob(directory)

for file in files:
        new_img = Image(file)
        new_img.show()
        time.sleep(1) #wait for 1 second

Download the script

As you can see it doesn’t take much to load up and start changing files.