mexican-government-report VS EasyOCR

Mexican Government Report Text Analysis using spaCy, NumPy, pandas, Matplotlib, Seaborn and geopandas.

EasyOCR is a popular project if you are in an environment where you can use run Python and PyTorch (https://github.com/JaidedAI/EasyOCR). Other open source projects of note are PaddleOCR (https://github.com/PaddlePaddle/PaddleOCR) and docTR (https://github.com/mindee/doctr).

The main one was https://github.com/JaidedAI/EasyOCR, mostly because, as promised, it was pretty easy to use, and uses pytorch (which I preferred in case I wanted to tweak it). It has been updated since, but at the time it was using CRNN, which is a solid model, especially for the time - it wasn't (academic) SOTA but not far behind that. I'm sure I could've coaxed better performance than I got out of it with some retraining and hyperparameter tuning.

I've used EasyOCR for number recognition tasks. Works fairly well. https://github.com/JaidedAI/EasyOCR

Do you have hardware acceleration available? If so, check out https://github.com/JaidedAI/EasyOCR

If all you need to do is OCR, check out https://github.com/JaidedAI/EasyOCR , it's a similar architecture to the cloud services, without all the $. You'll end up with extracted text and bounding boxes for it.

I've had good results with EasyOCR, much better than Tesseract. I agree with you, Tesseract has performed very poorly in my experience.

https://github.com/JaidedAI/EasyOCR

Currently the tool supports 2 different OCRs. PaddleOCR and EasyOCR. In both cases, the OCR has a specific model for Japanese characters. They work quite well, as long as the characters have clear contrast. The problem is that these OCR implement Torch, which makes the program very heavy. I am looking for a way to optimize this.

the repo with Python package https://github.com/JaidedAI/EasyOCR

website with online demo https://www.jaided.ai/easyocr/ (you can upload your images there)