Collecting Images for Training My Military AI: A Simple Approach

import requests
import os
import re

API_KEY = "API_KEY_HERE"
CX = "Search_Engine_Key"

def sanitize_folder_name(name):
    # Remove any characters that are not allowed in filenames or directories
    return re.sub(r'[<>:"/\\|?*]', '', name)

def search_images(query, num_images=100):
    print(f"Searching for: {query}")
    
    search_url = "https://www.googleapis.com/customsearch/v1"
    image_urls = []
    
    # Calculate how many requests we need to make (10 results per request)
    for start_index in range(1, num_images+1, 10):
        params = {
            "q": query,
            "cx": CX,
            "key": API_KEY,
            "searchType": "image",
            "num": 10,
            "start": start_index,  # Start from the next index
        }

        response = requests.get(search_url, params=params)
        
        print(f"Status code: {response.status_code}")
        try:
            data = response.json()
        except Exception as e:
            print("Error parsing JSON:", e)
            print("Raw response:", response.text)
            return

        # Extract image links inside the function
        items = data.get("items", [])
        image_urls.extend(

 for item in items])

        # If there are fewer than 10 items, stop searching further
        if len(items) < 10:
            break

    print("\nFound image URLs:")
    for url in image_urls:
        print(url)

    # Create a folder based on the query name
    folder_name = sanitize_folder_name(query)
    if not os.path.exists(folder_name):
        os.makedirs(folder_name)

    # Download images
    for i, url in enumerate(image_urls):
        try:
            response = requests.get(url)
            response.raise_for_status()  # Will raise an error if the response is not successful
            
            # Create the file path with the folder and image name
            file_name = os.path.join(folder_name, f"image_{i+1}.jpg")
            
            with open(file_name, 'wb') as file:
                file.write(response.content)
            
            print(f"Downloaded: {file_name}")
        except requests.exceptions.RequestException as e:
            print(f"Error downloading {url}: {e}")

# Call the function
search_images("BRDM-2 (Russia)", num_images=100)