HTML image extractor
Focused on the <img> tag: read src and split srcset into every candidate URL (descriptors like 2x or 640w dropped), optionally with alt text, lazy-load data-src and deduplication.
Runs in your browserEvery computation happens in your browser — your data never leaves this device.
Image links
One link per line, ready to copy.
Paste HTML to extract image links
img tags: 0 · Links: 0 · Unique: 0 · From srcset: 0
What this tool does
- Check whether a responsive srcset is missing a density or still points at an old domain.
- Export image URLs from a page for bulk download or a migration checklist.
- With data-src enabled, see which placeholders lazy-loaded images point at before scripts run.
Example
Input
<img src="a.png" srcset="a.png 1x, big.png 2x">
Output
a.png a.png big.png
Every srcset candidate is emitted and descriptors such as 2x are dropped.
Frequently asked questions
What happens to 2x and 640w descriptors?
They are dropped and only the URL is kept: a.png 1x, big.png 2x yields two lines, a.png and big.png, ready to download or compare.
Why does the same URL appear twice?
Because it is referenced by both src and srcset. Tick “Deduplicate by URL” to keep one line; the default keeps everything so you can see what each attribute points at.
Where is the lazy-load image URL?
Usually in an attribute like data-src or data-original, with a placeholder in src. Enable “Extract data-src” to include it.
Does it handle unclosed <img> tags?
Yes. img is a void element, so parsing stops at >; <img src=x> behaves the same as <img src="x" />.
Keywords：img 提取srcset 提取html 图片img src懒加载图片data-src响应式图片image src extractor