Files
douyi_guaji_main/operations/ocr_operations.py
T
Your Name 223c127e11 chore: Clean up and refactor codebase for improved maintainability
- Removed obsolete command-line test parameters from `run_cli.py` to streamline functionality.
- Introduced a new `ocr_preprocess.py` module for unified OCR image preprocessing, enhancing recognition stability.
- Updated various components to utilize the new preprocessing function, improving code organization and readability.
- Adjusted configuration values in `config.json` for better performance and consistency.
- Added comprehensive technical documentation for the Douyin project, detailing code modification standards and operational guidelines.
- Cleaned up excessive blank lines across multiple files to enhance overall readability.
2025-12-30 23:11:34 +08:00

207 lines
7.4 KiB
Python
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
"""OCR操作模块
包含OCR识别相关的操作。
"""
from __future__ import annotations
import os
import contextlib
from typing import Any, Optional
# PaddleOCR 3.x 会在初始化/导入时进行模型源连通性检查;在国内网络环境下容易卡住。
# 这里提前设置环境变量,避免启动耗时/卡死。
os.environ.setdefault("DISABLE_MODEL_SOURCE_CHECK", "True")
# 额外抑制 paddle/oneDNN/glog 的冗余信息(INFO/WARNING)
os.environ.setdefault("GLOG_minloglevel", "2")
os.environ.setdefault("OMP_NUM_THREADS", "1")
class OCRProcessor:
"""OCR处理器类"""
def __init__(self):
# 关闭文档方向/矫正等重型前处理:我们的截图是手机UI,不是扫描文档。
# 支持延迟初始化(测试流程/只做点击时不需要 OCR)。
self.ocr: Any = None
self.ocr_type = None # 'easyocr' 或 'paddleocr'
# 默认使用CPU模式(删除GPU检测)
self._gpu_available = False
if not str(os.environ.get("SKIP_OCR_INIT", "")).strip():
self._init_ocr()
def _init_ocr(self):
"""初始化OCR引擎"""
# 优先使用EasyOCR(更快),如果失败则回退到PaddleOCR
# 首先尝试EasyOCR
try:
# 在Windows上,避免使用stderr重定向,可能会导致"lost sys.stderr"问题
import easyocr # type: ignore[import-untyped, import] # pyright: ignore[reportMissingImports, reportMissingModuleSource]
# 只加载中文和英文,加快初始化速度
# 自动检测GPU可用性,如果可用则启用GPU加速
# 临时重定向输出以减少初始化时的噪音
import io
from contextlib import redirect_stdout, redirect_stderr
fake_stdout = io.StringIO()
fake_stderr = io.StringIO()
with redirect_stdout(fake_stdout), redirect_stderr(fake_stderr):
self.ocr = easyocr.Reader(
["ch_sim", "en"], gpu=False, verbose=False
)
self.ocr_type = "easyocr"
print(f"[OCR] 使用 EasyOCR(CPU模式)")
except Exception as e:
# EasyOCR初始化失败,回退到PaddleOCR
print(f"[OCR] EasyOCR初始化失败: {e},回退到PaddleOCR")
devnull = None
try:
devnull = open(os.devnull, "w")
with contextlib.redirect_stdout(
devnull
), contextlib.redirect_stderr(devnull):
from paddleocr import PaddleOCR
# 使用 PaddleOCR(强制 CPU 模式,不进行 GPU 检测)
use_gpu = False
self.ocr = PaddleOCR(
lang="ch",
ocr_version="PP-OCRv4",
use_angle_cls=False,
use_gpu=use_gpu,
use_doc_orientation_classify=False,
use_doc_unwarping=False,
use_textline_orientation=False,
)
self.ocr_type = "paddleocr"
print(f"[OCR] EasyOCR不可用,使用 PaddleOCR(CPU模式)")
print(
f"[OCR] 提示:安装EasyOCR可提升速度,运行: pip install easyocr"
)
except Exception:
# 若静默初始化失败,保留 None 并允许后续按需初始化,避免阻塞启动。
self.ocr = None
self.ocr_type = None
finally:
try:
if devnull is not None:
devnull.close()
except Exception:
pass
def _ensure_ocr(self):
"""确保OCR引擎已初始化"""
if self.ocr is not None:
return
# 默认使用CPU模式(延迟初始化时)
# 优先尝试EasyOCR
devnull = None
try:
devnull = open(os.devnull, "w")
with contextlib.redirect_stdout(devnull), contextlib.redirect_stderr(
devnull
):
import easyocr # type: ignore[import-untyped]
self.ocr = easyocr.Reader(
["ch_sim", "en"], gpu=False, verbose=False
)
self.ocr_type = "easyocr"
gpu_status = "CPU模式"
print(f"[OCR] 延迟初始化:使用 EasyOCR({gpu_status})")
except Exception:
# EasyOCR失败,回退到PaddleOCR
try:
if devnull is not None:
try:
devnull.close()
except Exception:
pass
except Exception:
pass
try:
devnull = open(os.devnull, "w")
with contextlib.redirect_stdout(devnull), contextlib.redirect_stderr(
devnull
):
from paddleocr import PaddleOCR
# 使用 PaddleOCR(强制 CPU 模式)
use_gpu = False
self.ocr = PaddleOCR(
lang="ch",
ocr_version="PP-OCRv4",
use_angle_cls=False,
use_gpu=use_gpu,
use_doc_orientation_classify=False,
use_doc_unwarping=False,
use_textline_orientation=False,
)
self.ocr_type = "paddleocr"
print(f"[OCR] 延迟初始化:EasyOCR不可用,使用 PaddleOCR(CPU模式)")
except Exception as e:
print(f"[OCR] OCR引擎初始化失败: {e}")
self.ocr = None
self.ocr_type = None
finally:
try:
if devnull is not None:
devnull.close()
except Exception:
pass
# 全局OCR处理器实例
_ocr_processor = OCRProcessor()
def _ensure_ocr():
"""确保OCR引擎已初始化(向后兼容接口)"""
_ocr_processor._ensure_ocr()
def analyse_pic_word(picname, change_color=0, pic_path=None):
"""文字OCR识别(向后兼容接口)"""
# 这里应该实现实际的OCR文字识别逻辑
# 暂时从原文件导入以保持功能
from Underlying_Operations import underlying_operations
instance = underlying_operations()
return instance.analyse_pic_word(picname, change_color, pic_path)
def analyse_pic_digits(picname, change_color=1, pic_path=None):
"""数字OCR识别(向后兼容接口)"""
# 这里应该实现实际的OCR数字识别逻辑
# 暂时从原文件导入以保持功能
from Underlying_Operations import underlying_operations
instance = underlying_operations()
return instance.analyse_pic_digits(picname, change_color, pic_path)
def extract_ocr_content(content=[], return_confidence=False):
"""提取OCR内容(向后兼容接口)"""
# 这里应该实现实际的OCR内容提取逻辑
# 暂时从原文件导入以保持功能
from Underlying_Operations import underlying_operations
instance = underlying_operations()
return instance.extract_ocr_content(content, return_confidence)
__all__ = [
"OCRProcessor",
"analyse_pic_word",
"analyse_pic_digits",
"extract_ocr_content",
"_ensure_ocr",
]