Amazon Listing Image Text Scraper
Pricing
from $2.24 / 1,000 images
Amazon Listing Image Text Scraper
Read text on public Amazon product photos. Each row is one gallery or A+ image with the transcribed text, claim lines, and spec pairs. Built for catalog and compliance teams. No login. Export JSON or CSV.
Pricing
from $2.24 / 1,000 images
Rating
0.0
(0)
Developer
rainminer
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
2 days ago
Last modified
Categories
Share
Read the words printed on Amazon product photos. Catalog and compliance teams use it when the claim, the size, or the ingredient line lives in the gallery or the A+ image and not in the bullet text.
Paste a public product URL such as https://www.amazon.com/dp/B07FZ8S74R. Each photo becomes one dataset row. No Amazon login is required.
What can this Amazon image text scraper do?
- Open a public Amazon product page.
- Collect gallery photos and A+ module images when the page includes them.
- Transcribe the text that is actually printed on each photo.
- Separate marketing claims from labeled spec pairs.
- Cap the number of photos per product with
maxItems. - Run several product URLs in one Actor run.
- Export JSON, CSV, or Excel and schedule a repeat through Apify.
maxItems counts photos per product URL, not a global total. A run with three products and maxItems of 3 can save up to nine rows.
What data can you extract?
| Field | Meaning |
|---|---|
| imageUrl | Photo that was read |
| imageRole | gallery or aplus |
| text | Visible text, in reading order |
| claims | Marketing sentences printed on the photo |
| specPairs | Labeled specs such as size or weight |
| asin | Amazon product id |
| productTitle | Title on the product page |
| productUrl | Product page you provided |
Empty text means that photo had no readable words. The row is still saved when the photo itself was read.
How to read Amazon listing images
- Open the product on Amazon and copy the URL. It should contain
/dp/and a 10-character ASIN. - Paste that URL into Start URLs. Add more products if you want a set.
- Set Maximum images. The default is 3 so a test stays small.
- Start the run and download the dataset.
Supported storefronts include amazon.com, amazon.co.uk, amazon.de, amazon.fr, amazon.it, amazon.es, amazon.ca, amazon.com.au, and amazon.co.jp. The ASIN in the URL is what the Actor keeps.
Use cases
- Check which claims are baked into the main image.
- Pull size and weight lines off an infographic.
- Compare A+ copy across a competitor set.
- Collect the words on a supplement or cosmetic facts panel when they are drawn on the photo.
- Review a new listing before the bullet text is translated.
- Sample three photos before reading a whole gallery.
- Build a sheet of ASINs with the on-image claims beside them.
- Watch one hero image on a schedule.
- Hand rows to a catalog QA workbook.
- Confirm a marketplace image still shows the same spec after a refresh.
Pricing
Platform usage is included.
| Charge | Free plan |
|---|---|
| Actor start | $0.00005 |
| Product page with a reading | $1.99 per 1,000 pages |
| Photo read | $2.99 per 1,000 photos |
Paid plans are cheaper. Keep maxItems small while you test a URL.
FAQ
Does this need an Amazon login? No. Public product pages are enough.
Why did my URL get skipped? The link must be an Amazon product URL with an ASIN, such as /dp/B07FZ8S74R.
Why is text null? That photo did not contain readable words. Claims and spec pairs are empty in the same case.
Image Credit
No site screenshot is used.
