googleImagesWebScraping

Helper To scrape Google images in order to build image library to feed tensorflow image recognition deep learning algorithm

prerequisits

install chrome web driver

go to:
https://chromedriver.storage.googleapis.com/index.html

installation is explained here:
https://www.youtube.com/watch?v=9kWz5tL4D7w

pip dependencies

pip install selenium
pip install urllib2
pip install BeautifulSoup4

usage

usage - in shell:

google_image_downloader.py -i <input_term_filepath> -o <output_image_folder_path> -t <theme> -n <nb_pictures_per_term> -d <web_driver_path>

-i or --input_term_filepath sets the path to the csv fiel containing all the keywords to scrap
-o or --output_image_folder_path set the path to the root folder where subfolder of terms will be downloaded
output folder structure will be as described below:
|---output_image_folder_path
____|---term 1
____|---term 2
____|---term 3
-t or --theme sets the general theme of the scraping : for instance "car", "cat", "food" in order to refine the search
-n --nb_pictures_per_term sets the number of images to download for each term
-d --web_driver_path sets the path to the chrome driver file