Google Images Scraper – Image Search Results

Scrape Google Images results as structured data. Type what you would type into Google's Images tab and every image comes back as a row: the full-size image address on the site that hosts it, its width and height, Google's thumbnail, the page the image appears on with its title and site name, the file size and the image's dominant colour — in Google's own order.

Built for dataset building, visual research, image SEO, product and brand monitoring, and finding the original source of a picture. Every filter from the Images tab's Tools menu is here: size, colour, type, file type, time and usage rights.

No API key, no login, no browser, no account of yours involved. About a hundred images arrive in one request, so a typical search finishes in seconds.

$0.50 per 1,000 images — $0.05 for a typical 100-image search. Apify platform usage included, nothing else to pay. A search that cannot be read is never charged.

Unofficial. Not affiliated with, endorsed by, or sponsored by Google.

Features

• The original, not the thumbnail — imageUrl is the full-size file on the site that hosts it, with its real pixel dimensions. Google's thumbnail is there too, in its own field.
• The page behind every image — pageUrl, its title, the site's name and domain, so you can check a licence or credit the source.
• Every Tools-menu filter — size, colour (including one specific colour or transparent backgrounds), type (photo, clip art, line drawing, animated GIF, face), file type, time, usage rights, and SafeSearch.
• An exact size floor — minWidth and minHeight keep only images at least that big, measured on Google's stated dimensions. Google's own size filter is coarser.
• Dominant colour and file size on every row — whatever filters you use.
• Many searches in one run — one per line, each with its own rows.
• Paste a Google Images link — a link copied from your browser is read with its own search terms and filters.
• Country and language — search as Google does from another country.
• Stable IDs — imageId is Google's own ID for the image, for deduplicating across searches and scheduled runs.
• Runs stay green. A search that cannot be read becomes an error row, so one bad line never costs the rest of the job.

Input

{
  "queries": ["golden gate bridge"],   // one search per line
  "maxItems": 100,                     // images per search (Google has ~300–400)
  "imageSize": "large",                // any | large | medium | icon
  "imageColor": "any",                 // any | color | grayscale | transparent | red | blue | …
  "imageType": "photo",                // any | photo | clipart | lineart | animated | face
  "fileType": "any",                   // any | jpg | png | gif | bmp | svg | webp | ico
  "timeRange": "any",                  // any | day | week | month | year
  "usageRights": "any",                // any | creativeCommons | commercial
  "safeSearch": false,
  "minWidth": 1200,                    // optional: drop anything narrower
  "country": "US",
  "language": "en"
}

Google's operators work in the search terms:

{
  "queries": [
    "site:nasa.gov mars rover",
    "\"red panda\" -cartoon",
    "https://www.google.com/search?q=eiffel+tower&udm=2&tbs=itp:animated"
  ]
}

Field	Type	Default	What it does
queries	array	—	Search terms, one per line, or links copied from Google's Images tab.
maxItems	integer	100	Images per search, up to 500. Google runs out at roughly 300–400 for most searches.
imageSize	string	any	large, medium or icon.
imageColor	string	any	color, grayscale, transparent, or one colour: red, orange, yellow, green, teal, blue, purple, pink, white, gray, black, brown.
imageType	string	any	photo, clipart, lineart, animated (GIFs) or face.
fileType	string	any	jpg, png, gif, bmp, svg, webp or ico.
timeRange	string	any	day, week, month or year.
usageRights	string	any	creativeCommons or commercial — Google's licence labels.
safeSearch	boolean	false	Have Google filter explicit images out.
minWidth, minHeight	integer	—	Drop images below this size in pixels. Exact, applied after the search.
country	string	US	Two-letter code: US, GB, DE, IN…
language	string	en	en, de, fr, es, pt-BR, ja…

A pasted Google link keeps its own filters; the filter fields apply to links that carry none, and to every plain search.

How many images

Google sends up to about a hundred images per request, so the default of 100 is usually a single request. Ask for more and further requests are made until you have them or Google runs out — for most searches that is somewhere around 300 to 400 images, the same point where the Images tab stops scrolling in a browser. Reading stops there on its own.

Output

One item per image, in the run's dataset:

{
  "type": "image",
  "position": 1,
  "title": "Red panda - Wikipedia",
  "imageUrl": "https://thumb.wikimedia.org/wikipedia/commons/thumb/f/fd/Red_Panda%2C_Gentle_Tree-Dweller_of_the_Himalayas.jpg/1280px-Red_Panda%2C_Gentle_Tree-Dweller_of_the_Himalayas.jpg?utm_source=en.wikipedia.org&utm_campaign=index&utm_content=thumbnail",
  "imageWidth": 1280,
  "imageHeight": 853,
  "thumbnailUrl": "https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcR4uYYm5fvNPU0Tzqjy66KfWvYLqXLkgsykJ5BNbGCntQ&s=10",
  "thumbnailWidth": 547,
  "thumbnailHeight": 365,
  "pageUrl": "https://en.wikipedia.org/wiki/Red_panda",
  "sourceName": "Wikipedia",
  "domain": "en.wikipedia.org",
  "fileSize": "269KB",
  "dominantColor": "#784e32",
  "imageId": "OqyF9koU4LIU_M",
  "query": "red panda",
  "searchUrl": "https://www.google.com/search?q=red+panda&udm=2&hl=en&gl=US",
  "page": 1,
  "retrievedAt": "2026-09-11T02:32:53.289Z"
}

Field	What it is
imageUrl	The full-size image on the site that hosts it — not Google's copy.
imageWidth, imageHeight	Full-size dimensions in pixels, as Google states them.
thumbnailUrl, thumbnailWidth, thumbnailHeight	Google's thumbnail, which stays up even if the original moves.
title	The title Google shows under the image — the title of the page it appears on.
pageUrl	The page the image appears on: where to check its licence or credit it.
sourceName, domain	The site's name as Google shows it, and its host.
fileSize	The full-size file's size, as Google states it.
dominantColor	The image's dominant colour, as a hex code.
imageId	Google's ID for the image, stable across runs — for deduplicating.
position, page	Google's rank, 1 first, and which request of the search it came from.
query, searchUrl	The search, and the Google Images address it was read from.

A search that could not be read is its own row with type: "error", an error code and an errorDescription. The run itself stays green either way, and its status message says how many searches could not be read — so if you run on a schedule, check for type: "error" rows rather than trusting the run status.

An empty dataset means the searches were read and Google showed no images for them. It never means a search could not be made; that arrives as an error row instead.

How to use

From Apify Console

1. Open the Actor and click Try for free / Start.
2. Put one search per line into Search terms — or paste a link from your browser's Google Images tab.
3. Set Images per search. 100 is one request; Google has roughly 300–400 for most searches.
4. Pick any Filters you want, exactly as in the Images tab's Tools menu. For images at least a certain size, set Minimum width or Minimum height.
5. Click Start, then open the Dataset tab — the Images view shows a thumbnail of each — and export as JSON, CSV or Excel.

From the API

curl -s "https://api.apify.com/v2/acts/simple.actor~google-images/run-sync-get-dataset-items?token=$APIFY_TOKEN" \
  -H 'Content-Type: application/json' \
  -d '{"queries": ["golden gate bridge"], "imageSize": "large", "maxItems": 100}'

With the JavaScript client:

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: process.env.APIFY_TOKEN });
const run = await client.actor('simple.actor/google-images').call({
    queries: ['red panda', 'snow leopard'],
    usageRights: 'creativeCommons',
    minWidth: 1600,
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();
// items[].imageUrl is the full-size file; items[].pageUrl is where to check its licence.

Use cases

• Training and evaluation datasets — collect hundreds of labelled images per term, with dimensions to filter on and a stable imageId to deduplicate.
• Image SEO — see which pages and sites rank in Google Images for your keywords, and where your own images sit.
• Brand and product monitoring — find where a logo or product photo appears, with the page that uses it.
• Source finding and licensing — go from a picture to the page that hosts it, and narrow to Creative Commons images before checking the licence.
• Content and design research — mood boards by colour, transparent PNGs, line drawings or GIFs for a topic.
• Local visual research — how a place or product looks in another country's results, with country and language.

Pricing

Simple pay-per-result pricing — no subscription and no start fee.

Event	Price
Image returned	$0.0005

What that comes to:

• A search at the default depth, 100 images: $0.05.
• Everything Google has for a search, roughly 350 images: about $0.18.
• 1,000 images, however many searches they come from: $0.50.

Error rows are never charged, and a search Google has no images for costs nothing.

Limits worth knowing

• Google stops at roughly 300–400 images a search. That is where the Images tab stops scrolling in a browser too; asking for more returns what Google has.
• Filters are Google's. Size, type, colour and licence are how Google has labelled the image. A licence label in particular is not a legal guarantee — check the licence on pageUrl before reusing an image.
• No aspect-ratio filter. Google currently ignores it, so it is not offered. Use imageWidth and imageHeight on the rows instead.
• A very specific or misspelled search is not flagged. Google loosens a search it has little for and shows related images without saying so.

FAQ

Is scraping Google Images legal? Results come from Google and are subject to Google's terms, and the images themselves belong to their owners. This Actor returns addresses and descriptions, not copies of the images; check the licence on each image's page before reusing it. Searches go through Apify's Google SERP proxy, which is the sanctioned route to Google's results pages.

Do I need a Google API key or a search API account? No. There is no key, no login, no browser and no quota of your own to manage.

Does it download the images? No. Each row carries the full-size address and Google's thumbnail; fetch them yourself, or feed imageUrl into a downloader.

How do I know a search failed if the run stays green? Check for rows with type: "error" — that is the failure channel by design, so one bad search never costs the rest of the job. The run's status message also counts how many searches could not be read.

Results come from Google and are subject to Google's terms. Searches use Apify's Google SERP proxy. Unofficial, and not affiliated with Google.