akaOCR Package Tools
Project description
akaOCR
Description
This package is compatible with akaOCR for ocr pipeline program (Text Detection & Text Recognition), using ONNX format model (CPU speed can be x2 times faster). This code is referenced from this awesome repo.
Installation
You can install the package using pip:
pip install akaocr
Usage
Step 1: Use in your code: Text Detection.
from akaocr import BoxEngine
import cv2
img_path = "path/to/image.jpg"
image = cv2.imread(img_path)
# side_len: minimum inference image size
box_engine = BoxEngine(model_path: Any | None = None,
side_len: int = 736,
conf_thres: float = 0.5)
# inference for one image
bboxs = box_engine(image) # [[4 points],...]
Step 2: Use in your code: Text Recognition.
from akaocr import TextEngine
import cv2
img_path = "path/to/cropped_image.jpg"
cropped_image = cv2.imread(img_path)
text_engine = TextEngine(model_path: Any | None = None,
vocab_path: Any | None = None,
use_space_char: bool = True)
# inference for one or more images
texts = text_engine(cropped_image) # [(text, conf),...]
License
This project is licensed under the Apache License 2.0
"I hope this package will be useful for your projects". Nguyễn Trường Lâu (from akaOCR Team).
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
akaocr-2.0.2.tar.gz
(10.4 MB
view hashes)
Built Distribution
akaocr-2.0.2-py3-none-any.whl
(10.4 MB
view hashes)