Hello trend,
This is Satya Mallick from LearnOpenCV.com.
In today's post we will discuss about Optical Character Recognition using PaddleOCR developed by the PaddlePaddle Team.
PaddleOCR is an OCR framework or toolkit that provides multilingual practical OCR tools that help the users to apply and train different models in a few lines of code. PaddleOCR offers a series of high-quality pretrained models for performing Text Detection, Text direction classification and Text Recognition.
You will learn the basics about algorithms like CRNN and the CTC Loss using in CRNN as well as perform OCR using the PP-OCRv3 models on different types of images like Receipts, Documents, License plates, ID Cards, Handwritten text, Road Signs, and Trading Cards. We also compare the speed and accuracy based on the Levenshtein distance metric between the supported models like SRN, and NRTR along with PP-OCRv2.
We found that PP-OCRv3 performs very fast and the accuracy is also better than other models available for off-the-shelf OCR inference.
Without further ado, let's get started.
| Optical Character Recognition using PaddleOCR |
You can check out the code by clicking on the button below, and star us on GitHub to say thanks!
| Download Code (GitHub) |
Cheers!
Satya
Courses / YouTube / Facebook / LinkedIn / Twitter / Instagram
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.