RapidOCR

import json
 
import logging
 
import sys
 
from rapidocr import RapidOCR
 
  
 
logging.disable(logging.INFO)
 
  
 
def main():
 
engine = RapidOCR()
 
  
 
img_url = sys.argv[1]
 
result = engine(img_url)
 
output = [{"text": txt, "box": box.tolist()} for txt, box in zip(result.txts, result.boxes)]
 
sys.stdout.buffer.write(json.dumps(output, ensure_ascii=False, indent=2).encode("utf-8"))
 
sys.stdout.buffer.write(b"\n")
 
  
 
if __name__ == "__main__":
 
main()

# -*- mode: python ; coding: utf-8 -*-
import sys
from PyInstaller.utils.hooks import collect_data_files, collect_submodules

# 包含模型文件
datas = collect_data_files("rapidocr")

hiddenimports = collect_submodules("rapidocr") + [
    "onnxruntime",
    "onnxruntime.capi._pybind_state",
]

# 按平台输出到不同目录: ocr-darwin / ocr-win32
platform_name = "darwin" if sys.platform == "darwin" else "win32"

a = Analysis(
    ["main.py"],
    pathex=[],
    binaries=[],
    datas=datas,
    hiddenimports=hiddenimports,
    hookspath=[],
    hooksconfig={},
    runtime_hooks=[],
    excludes=[
        "rapidocr_api",
        "fastapi",
        "uvicorn",
        "uvloop",
        "starlette",
    ],
    noarchive=False,
)

pyz = PYZ(a.pure)

exe = EXE(
    pyz,
    a.scripts,
    [],
    name="ocr",
    debug=False,
    bootloader_ignore_signals=False,
    strip=False,
    upx=True,
    upx_exclude=[],
    console=True,
    disable_windowed_traceback=False,
    argv_emulation=False,
    target_arch=None,
    codesign_identity=None,
    entitlements_file=None,
)

coll = COLLECT(
    exe,
    a.binaries,
    a.datas,
    strip=False,
    upx=True,
    upx_exclude=[],
    name=f"ocr-{platform_name}",
)