Frequently Asked Questions

How accurate is the extraction?

Accuracy depends on the resolution of your image and the thickness of the data lines. GraphScreen maps physical pixels to logical values. If a single pixel represents 0.5 units on your axis, your maximum theoretical precision is ±0.5 units. (Use our Error Margin tool to calculate this exactly).

Do you store my images or extracted data?

No. All processing happens locally in your browser using JavaScript. The images never touch our servers, making GraphScreen safe for proprietary financial or medical data.

Can it handle logarithmic scales?

Yes. During calibration, you can specify if the X or Y axis is Linear or Log (Base 10). If you are unsure, check our guide on identifying log scales.

Why not just use AI?

Current generation Large Multimodal Models (LMMs) do not physically "measure" pixels. They generate text that statistically correlates with the image. This results in hallucinated numbers. We prefer math over statistics when accuracy matters. Read more on our About page.

Can I process multiple lines at once?

Yes, you can create multiple "Datasets" within a single session to extract different colored lines independently, then export them as separate columns.