DEV Community

#ocr

Posts

👋 Sign in for the ability to sort posts by relevant, latest, or top.
Give Your Document OCR Pipeline a Memory

Give Your Document OCR Pipeline a Memory

Comments
5 min read
I built a free image preprocessor to get better OCR results (PrepOCR)

I built a free image preprocessor to get better OCR results (PrepOCR)

Comments
2 min read
OCR reads the invoice. The exception queue decides if it pays

OCR reads the invoice. The exception queue decides if it pays

Comments
7 min read
Designing an OCR pipeline: from scanned PDF to searchable, chunked text

Designing an OCR pipeline: from scanned PDF to searchable, chunked text

Comments
3 min read
The same pipeline, minus the cluster

The same pipeline, minus the cluster

1
Comments
5 min read
onError must return the right shape, or the crash just moves

onError must return the right shape, or the crash just moves

1
Comments
6 min read
Reading the results: text, boxes, entities and how they line up

Reading the results: text, boxes, entities and how they line up

1
Comments
6 min read
One bad page must not lose the other forty

One bad page must not lose the other forty

1
Comments
6 min read
Running OCR entirely in the browser meant decoupling detection from recognition

Running OCR entirely in the browser meant decoupling detection from recognition

1
Comments
8 min read
Privacy is an architecture, not a checkbox

Privacy is an architecture, not a checkbox

1
Comments
5 min read
A document pipeline in 30 lines, running in a browser tab

A document pipeline in 30 lines, running in a browser tab

1
Comments
6 min read
Your OCR pipeline probably uploads the document. It doesn't have to

Your OCR pipeline probably uploads the document. It doesn't have to

1
Comments
5 min read
Your OCR is turning smudges into empty cells, and you can't tell

Your OCR is turning smudges into empty cells, and you can't tell

Comments
5 min read
I Tested Three Vision Models on Catalog Images: OCR Was the Easy Part

I Tested Three Vision Models on Catalog Images: OCR Was the Easy Part

Comments
4 min read
Why we vendor 76 MB of OCR models instead of loading them from a CDN

Why we vendor 76 MB of OCR models instead of loading them from a CDN

Comments
8 min read
👋 Sign in for the ability to sort posts by relevant, latest, or top.