Skip to content

Python Script to download hundreds of images from 'Google Images'. It is a ready-to-run code!

License

Notifications You must be signed in to change notification settings

adayton1/google-images-download

 
 

Repository files navigation

Google Images Download

Python Script for 'searching' and 'downloading' hundreds of Google images to the local hard disk!

Contents

This is a command line python program to search keywords/key-phrases on Google Images and optionally download images to your computer. You can also invoke this script from another python file.

This is a small and ready-to-run program. No dependencies are required to be installed if you would only want to download up to 100 images per keyword. If you would want more than 100 images per keyword, then you would need to install Selenium library along with chromedriver. Detailed instructions in the troubleshooting section.

This program is compatible with both the versions of python - 2.x and 3.x (recommended). It is a download-and-run program with no changes to the file. You will just have to specify parameters through the command line.

You can use one of the below methods to download and use this repository.

Using pip

$ pip install google_images_download

Manually using CLI

$ git clone https://github.com/hardikvasa/google-images-download.git
$ cd google-images-download && sudo python setup.py install

Manually using UI

Go to the repo on github ==> Click on 'Clone or Download' ==> Click on 'Download ZIP' and save it on your local disk.

If installed via pip or using CLI, use the following command:

$ googleimagesdownload [Arguments...]

If downloaded via the UI, unzip the file downloaded, go to the 'google_images_download' directory and use one of the below commands:

$ python3 google_images_download.py [Arguments...]
OR
$ python google_images_download.py [Arguments...]

If you would want to use this library from another python file, you could use it as shown below:

from google_images_download import google_images_download

response = google_images_download.googleimagesdownload()
response.download({<Arguments...>})
Argument Short hand Description
config_file cf

You can pass the arguments inside a config file. This is an alternative to passing arguments on the command line directly.

Please refer to the config file format below

  • If 'config_file' argument is present, the program will use the config file and command line arguments will be discarded
  • Config file can only be in JSON format
  • Please refrain from passing invalid arguments from config file. Refer to the below arguments list
keywords k

Denotes the keywords/key phrases you want to search for. For more than one keywords, wrap it in single quotes.

Tips:

  • If you simply type the keyword, Google will best try to match it
  • If you want to search for exact phrase, you can wrap the keywords in double quotes ("")
  • If you want to search to contain either of the words provided, use OR between the words.
  • If you want to explicitly not want a specific word use a minus sign before the word (-)
keywords_from_file kf

Denotes the file name from where you would want to import the keywords.

Add one keyword per line. Blank/Empty lines are truncated automatically.

Only file types '.txt' or '.csv' are allowed.

prefix_keywords pk

Denotes additional words added before main keyword while making the search query.

The final search query would be: <prefix keyword> <keyword>

So, for example, if the keyword is 'car' and prefix_keyword is 'red,yellow,blue', it will search and download images for 'red car', 'yellow car' and 'blue car' individually

suffix_keywords sk

Denotes additional words added after main keyword while making the search query.

The final search query would be: <keyword> <suffix keyword>

So, for example, if the keyword is 'car' and suffix_keyword is 'red,yellow,blue', it will search and download images for 'car red', 'car yellow' and 'car blue' individually

limit l

Denotes number of images that you want to download.

You can specify any integer value here. It will try and get all the images that it finds in the google image search page.

If this value is not specified, it defaults to 100.

related_images ri

This argument downloads a ton of images related to the keyword you provided.

Google Images page returns list of related keywords to the keyword you have mentioned in the query. This tool downloads images from each of those related keywords based on the limit you have mentioned in your query

This argument does not take any value. Just add '--related_images' or '-ri' in your query.

Note: This argument can download hundreds or thousands of additional images so please use this carefully.

format f

Denotes the format/extension of the image that you want to download.

Possible values: jpg, gif, png, bmp, svg, webp, ico

color co

Denotes the color filter that you want to apply to the images.

Possible values: red, orange, yellow, green, teal, blue, purple, pink, white, gray, black, brown

color_type ct

Denotes the color type you want to apply to the images.

Possible values: full-color, black-and-white, transparent

usage_rights r

Denotes the usage rights/licence under which the image is classified.

Possible values:

  • labeled-for-reuse-with-modifications,
  • labeled-for-reuse,
  • labeled-for-noncommercial-reuse-with-modification,
  • labeled-for-nocommercial-reuse
size s

Denotes the relative size of the image to be downloaded.

Possible values: large, medium, icon, >400*300, >640*480, >800*600, >1024*768, >2MP, >4MP, >6MP, >8MP, >10MP, >12MP, >15MP, >20MP, >40MP, >70MP

aspect_ratio a

Denotes the aspect ratio of images to download.

Possible values: tall, square, wide, panoramic

type t

Denotes the type of image to be downloaded.

Possible values: face, photo, clip-art, line-drawing, animated

time w

Denotes the time the image was uploaded/indexed.

Possible values: past-24-hours, past-7-days

time_range wr

Denotes the time range for which you want to search the images

The value of this parameter should be in the following format '{"time_min":"MM/DD/YYYY","time_max":"MM/DD/YYYY"}'

delay d

Time to wait between downloading two images

Time is to be specified in seconds. But you can have sub-second times by using decimal points.

url u

Allows you search by image when you have the URL from the Google Images page. It downloads images from the google images link provided

If you are searching an image on the browser google images page, simply grab the browser URL and paste it in this parameter It will download all the images seen on that page.

single_image x Allows you to download one image if the complete (absolute) URL of the image is provided
output_directory o

Allows you specify the main directory name in which the images are downloaded.

If not specified, it will default to 'downloads' directory. This directory is located in the path from where you run this code

proxy px

Allows you to specify proxy server setting for all your requests

You can specify the proxy settings in 'IP:Port' format

similar_images si

Reverse Image Search or 'Search by Image' as it is referred to on Google.

Searches and downloads images that are similar to the absolute image link/url you provide.

specific_site ss Allows you to download images with keywords only from a specific website/domain name you mention.
print_urls p

Print the URLs of the images on the console. These image URLs can be used for debugging purposes

This argument does not take any value. Just add '--print_urls' or '-p' in your query.

print_size ps

Prints the size of the images on the console

The size denoted the actual size of the image and not the size of the image on disk

This argument does not take any value. Just add '--print_size' or '-ps' in your query.

metadata m

Prints the metada of the image on the console.

This includes image size, origin, image attributes, description, image URL, etc.

This argument does not take any value. Just add '--metadata' or '-m' in your query.

extract_metadata e

This option allows you to save metadata of all the downloaded images in a text file.

This file can be found in the logs/ directory. The name of the file would be same as the keyword nam

This argument does not take any value. Just add '--extract_metadata' or '-e' in your query.

socket_timeout st

Allows you to specify the time to wait for socket connection.

You could specify a higher timeout time for slow internet connection. The default value is 10 seconds.

thumbnail th

Downloads image thumbnails corresponding to each image downloaded.

Thumbnails are saved in their own sub-directories inside of the main directory.

This argument does not take any value. Just add '--thumbnail' or '-th' in your query.

language la

Defines the language filter. The search results are automatically returned in that language

Possible Values: Arabic, Chinese (Simplified), Chinese (Traditional), Czech, Danish, Dutch, English, Estonian. Finnish, French, German, Greek, Hebrew, Hungarian, Icelandic, Italian, Japanese, Korean, Latvianm, Lithuanian, Norwegian, Portuguese, Polish, Romanian, Russian, Spanish, Swedish, Turkish

prefix pr

A word that you would want to prefix in front of actual image name.

This feature can be used to rename files for image identification purpose.

chromedriver cd

With this argument you can pass the path to the 'chromedriver'.

The path looks like this: "path/to/chromedriver". In windows it will be "path/to/chromedriver.exe"

help h show the help message regarding the usage of the above arguments

Note: If single_image or url parameter is not present, then keywords is a mandatory parameter. No other parameters are mandatory.

You can either pass the arguments directly from the command as in the examples below or you can pass it through a config file. Below is a sample of how a config file looks.

You can pass more than one record through a config file. The below sample consist of two set of records. The code will iterate through each of the record and download images based on arguments passed.

{
    "Records": [
        {
            "keywords": "apple",
            "limit": 5,
            "color": "green",
            "print_urls": true
        },
        {
            "keywords": "universe",
            "limit": 15,
            "size": "large",
            "print_urls": true
        }
    ]
}
  • If you are calling this library from another python file, below is the sample code
from google_images_download import google_images_download   #importing the library

response = google_images_download.googleimagesdownload()   #class instantiation

arguments = {"keywords":"Polar bears,baloons,Beaches","limit":20,"print_urls":True}   #creating list of arguments
response.download(arguments)   #passing the arguments to the function
  • If you are passing arguments from a config file, simply pass the config_file argument with name of your JSON file
$ googleimagesdownload -cf example.json
  • Simple example of using keywords and limit arguments
$ googleimagesdownload --keywords "Polar bears, baloons, Beaches" --limit 20
  • Using Suffix Keywords allows you to specify words after the main keywords. For example if the keyword = car and suffix keyword = 'red,blue' then it will first search for car red and then car blue
$ googleimagesdownload --k "car" -sk 'red,blue,white' -l 10
  • To use the short hand command
$ googleimagesdownload -k "Polar bears, baloons, Beaches" -l 20
  • To download images with specific image extension/format
$ googleimagesdownload --keywords "logo" --format svg
  • To use color filters for the images
$ googleimagesdownload -k "playground" -l 20 -co red
  • To use non-English keywords for image search
$ googleimagesdownload -k "北极熊" -l 5
  • To download images from the google images link
$ googleimagesdownload -k "sample" -u <google images page URL>
  • To save images in specific main directory (instead of in 'downloads')
$ googleimagesdownload -k "boat" -o "boat_new"
  • To download one single image with the image URL
$ googleimagesdownload --keywords "baloons" --single_image <URL of the images>
  • To download images with size and type constrains
$ googleimagesdownload --keywords "baloons" --size medium --type animated
  • To download images with specific usage rights
$ googleimagesdownload --keywords "universe" --usage_rights labeled-for-reuse
  • To download images with specific color type
$ googleimagesdownload --keywords "flowers" --color_type black-and-white
  • To download images with specific aspect ratio
$ googleimagesdownload --keywords "universe" --aspect_ratio panoramic
  • To download images which are similar to the image in the image URL that you provided (Reverse Image search).
$ googleimagesdownload -si <image url> -l 10
  • To download images from specific website or domain name for a given keyword
$ googleimagesdownload --keywords "universe" --specific_site example.com

===> The images would be downloaded in their own sub-directories inside the main directory (either the one you provided or in 'downloads') in the same folder you are in.


## SSL Errors

If you do see SSL errors on Mac for Python 3, please go to Finder —> Applications —> Python 3 —> Click on the ‘Install Certificates.command’ and run the file.

## googleimagesdownload: command not found

While using the above commands, if you get Error: -bash: googleimagesdownload: command not found then you have to set the correct path variable.

To get the details of the repo, run the following command:

$ pip show -f google_images_download

you will get the result like this:

Location: /Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages
Files:
  ../../../bin/googleimagesdownload

together they make: /Library/Frameworks/Python.framework/Versions/2.7/bin which you need add it to the path:

$ export PATH="/Library/Frameworks/Python.framework/Versions/2.7/bin"

## [Errno 13] Permission denied creating directory 'downloads'

When you run the command, it downloads the images in the current directory (the directory from where you are running the command). If you get permission denied error for creating the downloads directory, then move to a directory in which you have the write permission and then run the command again.

## Permission denied while installing the library

On MAC and Linux, when you get permission denied when installing the library using pip, try doing a user install.

$ pip install google_images_download --user

You can also run pip install as a superuser with sudo pip install google_images_download but it is not generally a good idea because it can cause issues with your system-level packages.

## Installing the chromedriver (with Selenium)

If you would want to download more than 100 images per keyword, then you will need to install 'selenium' along with 'chromedriver'.

If you have pip installed the library or run the setup.py file, Selenium would have automatically installed on your machine. You will also need Chrome browser on your machine. For chromedriver:

Download the correct chromedriver based on your operating system.

On Windows or MAC if for some reason the chromedriver gives you trouble, download it under the current directory and run the command.

On Linux if you are having issues installing google chrome browser, refer to this CentOS or Amazon Linux Guide or Ubuntu Guide

For All the operating systems you will have to use '--chromedriver' or '-cd' argument to specify the path of chromedriver that you have downloaded in your machine.

Below diagram represents the code logic.

Anyone is welcomed to contribute to this script. If you would like to make a change, open a pull request. For issues and discussion visit the Issue Tracker.

The aim of this repo is to keep it simple, stand-alone, backward compatible and 3rd party dependency proof.

This program lets you download tons of images from Google. Please do not download or use any image that violates its copyright terms. Google Images is a search engine that merely indexes images and allows you to find them. It does NOT produce its own images and, as such, it doesn't own copyright on any of them. The original creators of the images own the copyrights.

Images published in the United States are automatically copyrighted by their owners, even if they do not explicitly carry a copyright warning. You may not reproduce copyright images without their owner's permission, except in "fair use" cases, or you could risk running into lawyer's warnings, cease-and-desist letters, and copyright suits. Please be very careful before its usage!

About

Python Script to download hundreds of images from 'Google Images'. It is a ready-to-run code!

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%