Get the text out of the picture.

Receipts, labels, forms, signs, screenshots — anywhere the information you need is trapped in an image.

What text appears in this image?

Images.

What makes this hard by hand.

The data exists but isn't usable
Information locked in a photo can't be searched, filtered, or reported on until something reads it.
Re-keying by hand is slow and error-prone
And it scales linearly with volume, which is the wrong shape.
Often step one, not the whole job
Read the text, then classify or route on what it says.

Three steps.

1

Create an OCR function

Choose OCR as the output modality when you create the function. Image input.

2

Send images, get text back

Same invoke pattern as any other function type.

3

Chain it

Feed the extracted text into a classification function to sort, route, or flag on its contents.