Check Your Website for Copyrighted Images

List all the images your site publishes, then check each one for ownership signals: watermarks, copyright metadata and matches to images published elsewhere. For every flagged image, find the license, receipt or written permission that covers your use. If you can't, license it, replace it or remove it, and record what you decided.

1. List the images your pages actually publish. Crawl the site or export your media library. Include the images people forget: theme demo images, sliders, background images, social share images and images inside PDFs.
2. Decide what to check first. Start with pages that earn money or get the most traffic, images a former designer or agency supplied, older posts from before you tracked licenses, and anything you can't trace to a receipt.
3. Check each image for ownership signals. Look for watermarks and credit lines, read the copyright metadata, and run a reverse image search to see where else the image appears.
4. Match flagged images to your records. A license only helps if it covers this use. Check whose account bought it, the license type (commercial or editorial only) and any limits.
5. Keep, license, replace or remove. If you remove an image, remove every copy, including resized versions and cached files. Write down what you did and when.
6. Check again after changes. Scan again after a redesign, a site migration or a batch of new content.

PixGuard does most of the finding and checking in steps 1 and 3 for the pages it can crawl (its version of a reverse image search, source lookup, is on Pro and Business), and its risk ranking gives you a starting order for step 2. Steps 4 and 5 need your records, which no scanner can see. For the full method, including an inventory spreadsheet and a triage order, read our guide to checking your website's images for copyright or follow the step-by-step website image audit.

What runs depends on how you scan. The no-signup check on this page runs a lighter set of checks than a scan from a free account, and Pro and Business add matching and lookup features.

The no-signup check and account scans use different pipelines, so their scores aren't directly comparable. Every check is probabilistic: any of them can miss an image, and any of them can flag an image you are entitled to use.

The free page check (no account)

1. You enter the address of one web page.
2. PixGuard downloads that page's HTML (without running JavaScript), looks at the first 30 images it finds and picks the 3 largest.
3. It checks those 3 images for watermark patterns, stock-photo style and unusual pixel patterns, and gives each one a 0 to 100 score with up to 5 signal labels.
4. Results usually appear in under a minute. They aren't saved to an account, and you can run 3 checks a day.

If the check finds no images, the page probably loads them with JavaScript. Try another page, such as a blog post or a product page.

A website scan (free account)

1. Enter your site's address in the dashboard. Use the exact address your site uses, with or without www, because the crawler stays on that host. Subdomains aren't included.
2. Before the scan starts, PixGuard estimates how many images it will find and how many image scans that will use.
3. The crawler follows links from that address across up to 50 pages and collects the images it finds.
4. Each new image uses 1 image scan (1 credit). Images PixGuard has already analyzed are shown from cache at no cost. Your plan sets how many new images one scan analyzes: 10 on Free, 500 on Pro and 2,000 on Business.
5. Results are sorted by risk, with High, Medium and Low counts at the top. Open any image to see its score, a plain-English summary and the signals behind it. On Pro and Business, high-risk images also show source lookup results.
6. Images that can't be analyzed are marked and not charged, and you can finish them later. PixGuard emails you when the scan is done.

A 10-image free scan usually takes about a minute. A scan of several hundred images can take 30 minutes or more.

Single images and PDFs

Upload one image from your dashboard (JPG, PNG, WebP or GIF, up to 10 MB) and it gets the same checks as an account website scan on your plan. On Pro and Business you can also upload a PDF, such as a brochure or catalog, and check the images inside it (up to 25 MB and 50 images per file).

A PixGuard score measures how strongly an image shows copyright risk signals. It isn't a legal finding, and it can't see your licenses. Use it to decide what to investigate first.

In a website scan's results list, the badges group these bands: Safe and Low show as Safe, Medium as Needs Review, and High and Critical as Risky.

A high score can be fine: you may have bought that stock photo, and the agency metadata or fingerprint match that raised the score simply shows where it came from. A visible stock watermark is different: licensed downloads don't carry one, so a watermarked preview on a live page usually means the license was never bought. A low score can be wrong: an image with stripped metadata and no watermark can still belong to someone.

The free page check uses the same 0 to 100 scale and bands, applied to its lighter set of checks. An image with no signal labels and a score under 20 shows "No risk signals found".

To see these bands on real results, start with the page that matters most, such as your homepage or a best-selling product page: Find Images to Investigate (free, no signup, its 3 largest images).

Google Images is a search engine. It shows images hosted on other websites, and those images still belong to whoever created or licensed them. Finding an image through Google gives you no permission to use it.

The Usage rights filter (Tools, then Usage rights) narrows results to images whose pages say they are available under Creative Commons or commercial licenses. The filter relies on what the hosting page claims, so open the source page, confirm the license and follow its conditions, such as attribution, before you publish. A "Licensable" badge means the site or image provider supplied license information with the image, not that the image is free.

Crediting the photographer doesn't create permission either. Attribution only satisfies a license that already allows your use, such as many Creative Commons licenses.

More detail: Can I use Google Images on my website?

Image files can carry text fields about who made them and on what terms. EXIF has Copyright and Artist fields. IPTC has Creator, Copyright Notice, Credit Line, Web Statement of Rights and Licensor URL. XMP carries the same kind of information for editing workflows. Some files also carry C2PA Content Credentials, a signed record of where an image came from and how it was edited.

What metadata can tell you: who claims to own the image, which agency or photographer supplied it, and sometimes where to license it. An agency name in the Credit Line is a strong lead.

What it cannot tell you:

• Whether you have a license. Metadata describes the image, not your purchase.
• Anything when it's missing. Many tools strip metadata when they resize or recompress a file. WordPress keeps it in resized copies only when the host uses the Imagick editor rather than GD, and Cloudflare Images keeps only the EXIF copyright tag by default. Empty fields prove nothing.
• The truth, reliably. Anyone can edit these fields, so a name in them is a claim, not proof.

Don't strip copyright information from the images you use. In the US, intentionally removing or altering copyright management information, such as a creator or copyright field, when you know or have reason to know it will help conceal an infringement, can be a separate violation (17 U.S.C. §1202).

Account scans read the copyright, creator and rights fields automatically. To see the main copyright, creator and rights fields in one image, use the free image metadata viewer. Our guide to EXIF and IPTC copyright fields explains what each one means.

A reverse image search, such as the visual search in Google or Bing, finds pages where the same or a similar image appears.

What it can tell you: where else the image is published, whether it's sold on a stock site, and which page is the likely original source. That leads you to the owner and the license terms. Our guide shows how to find out who owns an image.

What it cannot tell you:

• Whether you're licensed. A match on a stock site means the image can be licensed. You may already have bought it.
• That an image is free when nothing matches. New, unindexed or heavily edited images often return no results.
• A complete answer. Each engine indexes different pages, so results vary. Try more than one.

On Pro and Business, PixGuard's source lookup shows web pages where a matching image appears for your high-risk images, so you don't have to search each one by hand. Reverse image search for copyright checking covers the manual method.

A flag means "check this one". Work through it in this order:

1. Read the evidence. Open the image's signals to see what raised the score: a watermark, a metadata field or a fingerprint match.
2. Search your records. Look for the license, invoice or written permission. Check stock accounts in your business's name and in the names of anyone who built the site, such as a former designer or agency.
3. If you hold a license, confirm it covers this use. Check the license type (editorial licenses usually exclude commercial use), the licensee name and any limits. File a copy with the page URL. On paid plans you can mark the image as Licensed in PixGuard's Risk Inbox. See what Shutterstock and Adobe Stock licenses allow.
4. If you can't find one, fix it. License the image from its owner, replace it with one you own or can prove you licensed, or remove it. Remove every copy, including thumbnails, resized versions and CDN caches. Our guide covers how to replace copyrighted images safely.
5. Record the decision. Note the image, the page, what you did and the date. PixGuard doesn't export results, so keep this log in your own records.

Already received a demand letter or a takedown notice? Don't ignore it, and don't pay before you've checked the claim. Removing the image stops future use but doesn't by itself settle a claim about past use. Our guides cover Getty Images demand letters, DMCA takedown notices and the US Copyright Claims Board. For advice on a specific claim, talk to a copyright lawyer in your country.

Limits per scan

• Pages: up to 50 per scan, following links from the address you enter, on the same host.
• New images per scan: 10 on Free, 500 on Pro and 2,000 on Business. Each new image uses 1 image scan from your allowance.
• Repeat images: images PixGuard has already analyzed are shown from cache at no cost, so a repeat scan mostly spends image scans on new images.
• Allowances: Free includes 30 image scans once, and they expire 30 days after signup. Pro includes about 250 images a month and Business about 1,000. See pricing.
• Sites larger than 50 pages:start further scans from other sections, such as your blog index or a product category, to reach deeper pages. Images already analyzed aren't charged again.

What the crawler can miss

The crawler reads each page's HTML the way a simple bot does. It finds images in <img> tags and background images set in a page's inline styles. A lazy-loaded image is found when its real address is in src, or in data-src with an empty src; many lazy-loading plugins put a placeholder in src instead, and those images are missed. The crawler does not see:

• images added by JavaScript after the page loads, such as some sliders, galleries and page-builder blocks;
• alternative sizes listed in srcset or <picture> elements;
• background images defined in stylesheets;
• social share images (og:image) and favicons;
• images inside PDFs or other documents you link to;
• pages that no crawled page links to, pages behind a login, and subdomains;
• pages behind a bot challenge. If your firewall blocks the scan, allowlist the PixGuardBot user agent.

Results list each image's URL but not the page it was found on. To find where a flagged image is used, search your content management system or media library for its file name.

To cover the gaps, upload PDFs on Pro and Business, upload individual images from your dashboard, and check the images your theme sets in its stylesheet by hand.

Rescans and monitoring

Scan again after a redesign, a site migration, a new theme or a batch of new posts. Only new images use image scans. Pro can schedule weekly re-scans of up to 3 sites, and Business daily or weekly re-scans of up to 10 sites, each with an email summary of risky images.

WordPress sites collect images from many sources: theme demo content, page-builder templates, plugins, guest authors and years of uploads.

• The free page check and account website scans work on public WordPress sites, with the limits above.
• A URL crawl only sees images on pages it can reach. Media Library uploads that aren't on a linked page, and images a slider or gallery loads with JavaScript, need another route, such as uploading them from your dashboard.
• The PixGuard Scanner plugin is free to install from the WordPress.org plugin directory. It connects to your PixGuard account with an API key from Settings, and its scans use your account's image scans, 1 per new image.
• Theme and page-builder demo images are sometimes licensed for the demo only. If you kept them after launch, check the theme's license notes.

See how PixGuard works with WordPress, including the free PixGuard Scanner plugin. Scans use your account's image scans.

Stores usually have three groups of images worth checking first:

• Product photos from suppliers or manufacturers.Permission to sell a product doesn't always include permission to reuse its photos. Check your supplier or dropshipping agreement.
• Lifestyle and banner images on the homepage and collection pages, often from stock sites or a former designer.
• Theme demo images left in place after launch.

A site scan covers up to 50 pages, so start from your homepage and then from your largest category pages to reach more products. Store themes often load product galleries with JavaScript; if a scan finds fewer images than you expect, upload the key product images from your dashboard. Marketplace listings on Amazon or Etsy live on those platforms' domains, so a scan of your own store won't reach them.

More for stores: image copyright for online stores, Shopify image copyright and copyright risks for Amazon sellers.

If you build or manage sites for clients, check images before launch and at handover, and record where each image came from and whose account holds the license. A license bought in an agency's or freelancer's name may not cover the client. Read more on who is liable when an agency uses an unlicensed image.

What PixGuard offers for this today:

• Business: about 1,000 images a month, up to 2,000 new images per scan, scheduled re-scans of up to 10 sites (daily or weekly) and API access.
• Pro: about 250 images a month, up to 500 new images per scan and weekly re-scans of up to 3 sites.
• API: scan images by upload or URL, fetch reports and manage monitoring with an API key. See the API reference.
• Limits to plan around:one scan runs at a time per account and later scans queue; there is no CSV or PDF export, and results don't record the page each image was found on. Keep client records in your own audit log.

Maintained by Suraj, founder of PixGuard.

The features, limits and plan allowances on this page come from PixGuard's own scanning code and plan settings, checked in September 2026. When they change, we update this page.

Legal points refer to US law (Title 17 of the US Code) unless stated otherwise. Rules differ in other countries. This page is general information, not legal advice. For a specific claim, talk to a copyright lawyer where you are.

Last updated: September 22, 2026