Skip to main content

影刀离线OCR

Project description

简介

利用 onnxruntime 及 PaddleOCR 提供的模型, 对图片中的文字进行检测与识别.

使用模型

  • 文字检测: ch_PP-OCRv3_det_infer
  • 方向分类: cls mobile v2
  • 文字识别: ch_PP-OCRv2_rec_infer

参考

安装

pip install ydocr

使用

import cv2
from ydocr.predict_system import TextSystem,order_onrow

text_sys = TextSystem()

# 识别单行文本
res = text_sys.ocr_single_line(cv2.imread('single_line_text.png'))
print(res)

# 批量识别单行文本
res = text_sys.ocr_lines([cv2.imread('single_line_text.png')])
print(res[0])

# 检测并识别文本
img = cv2.imread('test.png')
res = text_sys.detect_and_ocr(img)
for boxed_result in res:
    print("{}, {:.3f}".format(boxed_result.ocr_text, boxed_result.score))


# 检测并识别文本(换行执行)
from ydocr.predict_system import TextSystem,order_onrow
img = cv2.imread('test.png')
res = text_sys.detect_and_ocr(img)
res = order_onrow(res)

#检测并识别文本,图片源是网络图片
from urllib.request import urlopen
def cv_imread_url(img_url):
    resp = urlopen(img_url)
    img = np.asarray(bytearray(resp.read()), dtype=np.uint8)
    img = cv2.imdecode(img, cv2.IMREAD_COLOR)
    return img 

img = cv_imread_url('https://imgurl.png')

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

ydocr-1.2.8.tar.gz (17.6 MB view hashes)

Uploaded Source

Built Distribution

ydocr-1.2.8-py3-none-any.whl (17.7 MB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page