WeKnora/docreader/parser/ocr_engine.py

328 lines
11 KiB
Python
Raw Normal View History

import io
2025-08-05 07:08:07 +00:00
import logging
import os
import platform
import subprocess
2025-08-05 07:08:07 +00:00
from abc import ABC, abstractmethod
from typing import Dict, Union
2025-08-05 07:08:07 +00:00
import numpy as np
from openai import OpenAI
from PIL import Image
from docreader.utils import endecode
2025-08-05 07:08:07 +00:00
logger = logging.getLogger(__name__)
2025-08-05 07:08:07 +00:00
class OCRBackend(ABC):
"""Base class for OCR backends"""
2025-08-05 07:08:07 +00:00
@abstractmethod
def predict(self, image: Union[str, bytes, Image.Image]) -> str:
"""Extract text from an image
2025-08-05 07:08:07 +00:00
Args:
image: Image file path, bytes, or PIL Image object
2025-08-05 07:08:07 +00:00
Returns:
Extracted text
"""
pass
class DummyOCRBackend(OCRBackend):
"""Dummy OCR backend implementation"""
def predict(self, image: Union[str, bytes, Image.Image]) -> str:
logger.warning("Dummy OCR backend is used")
return ""
2025-08-05 07:08:07 +00:00
class PaddleOCRBackend(OCRBackend):
"""PaddleOCR backend implementation"""
def __init__(self):
2025-08-05 07:08:07 +00:00
"""Initialize PaddleOCR backend"""
self.ocr = None
try:
2025-09-09 16:43:46 +00:00
import paddle
2025-09-09 16:43:46 +00:00
# Set PaddlePaddle to use CPU and disable GPU
os.environ["CUDA_VISIBLE_DEVICES"] = ""
paddle.device.set_device("cpu")
2025-09-11 03:45:34 +00:00
# 尝试检测CPU是否支持AVX指令集
try:
# 检测CPU是否支持AVX
if platform.system() == "Linux":
try:
result = subprocess.run(
["grep", "-o", "avx", "/proc/cpuinfo"],
capture_output=True,
text=True,
timeout=5,
)
has_avx = "avx" in result.stdout.lower()
2025-09-11 03:45:34 +00:00
if not has_avx:
logger.warning(
"CPU does not support AVX instructions, "
"using compatibility mode"
)
2025-09-11 03:45:34 +00:00
# 进一步限制指令集使用
os.environ["FLAGS_use_avx2"] = "0"
os.environ["FLAGS_use_avx"] = "1"
except (
subprocess.TimeoutExpired,
FileNotFoundError,
subprocess.SubprocessError,
):
logger.warning(
"Could not detect AVX support, using compatibility mode"
)
os.environ["FLAGS_use_avx2"] = "0"
os.environ["FLAGS_use_avx"] = "1"
2025-09-11 03:45:34 +00:00
except Exception as e:
logger.warning(
f"Error detecting CPU capabilities: {e}, using compatibility mode"
)
os.environ["FLAGS_use_avx2"] = "0"
os.environ["FLAGS_use_avx"] = "1"
2025-08-05 07:08:07 +00:00
from paddleocr import PaddleOCR
# OCR configuration with text orientation classification enabled
2025-08-05 07:08:07 +00:00
ocr_config = {
2025-09-09 16:43:46 +00:00
"use_gpu": False,
"text_det_limit_type": "max",
"text_det_limit_side_len": 960,
"use_doc_orientation_classify": True, # 启用文档方向分类
2025-09-09 16:43:46 +00:00
"use_doc_unwarping": False,
"use_textline_orientation": True, # 启用文本行方向检测
2025-09-09 16:43:46 +00:00
"text_recognition_model_name": "PP-OCRv4_server_rec",
"text_detection_model_name": "PP-OCRv4_server_det",
"text_det_thresh": 0.3,
"text_det_box_thresh": 0.6,
"text_det_unclip_ratio": 1.5,
"text_rec_score_thresh": 0.0,
"ocr_version": "PP-OCRv4",
2025-08-05 07:08:07 +00:00
"lang": "ch",
2025-09-09 16:43:46 +00:00
"show_log": False,
"use_dilation": True, # improves accuracy
"det_db_score_mode": "slow", # improves accuracy
2025-08-05 07:08:07 +00:00
}
2025-08-05 07:08:07 +00:00
self.ocr = PaddleOCR(**ocr_config)
logger.info("PaddleOCR engine initialized successfully")
2025-09-09 16:43:46 +00:00
except ImportError as e:
logger.error(
f"Failed to import paddleocr: {str(e)}. "
"Please install it with 'pip install paddleocr'"
)
2025-09-11 03:45:34 +00:00
except OSError as e:
if "Illegal instruction" in str(e) or "core dumped" in str(e):
logger.error(
f"PaddlePaddle crashed due to CPU instruction set incompatibility:"
f"{e}"
)
logger.error(
"This happens when the CPU doesn't support AVX instructions. "
"Try install CPU-only version of PaddlePaddle, "
"or use a different OCR backend."
)
2025-09-11 03:45:34 +00:00
else:
logger.error(
f"Failed to initialize PaddleOCR due to OS error: {str(e)}"
)
2025-08-05 07:08:07 +00:00
except Exception as e:
logger.error(f"Failed to initialize PaddleOCR: {str(e)}")
def predict(self, image: Union[str, bytes, Image.Image]) -> str:
"""Extract text from an image
Args:
image: Image file path, bytes, or PIL Image object
Returns:
Extracted text
"""
if isinstance(image, str):
image = Image.open(image)
elif isinstance(image, bytes):
image = Image.open(io.BytesIO(image))
if not isinstance(image, Image.Image):
raise TypeError("image must be a string, bytes, or PIL Image object")
return self._predict(image)
def _predict(self, image: Image.Image) -> str:
2025-08-05 07:08:07 +00:00
"""Perform OCR recognition on the image
Args:
image: Image object (PIL.Image or numpy array)
Returns:
Extracted text string
"""
if self.ocr is None:
logger.error("PaddleOCR engine not initialized")
return ""
2025-08-05 07:08:07 +00:00
try:
# Ensure image is in RGB format
if image.mode != "RGB":
2025-09-09 16:43:46 +00:00
image = image.convert("RGB")
2025-08-05 07:08:07 +00:00
2025-09-09 16:43:46 +00:00
# Convert to numpy array if needed
image_array = np.array(image)
2025-08-05 07:08:07 +00:00
2025-09-09 16:43:46 +00:00
# Perform OCR
ocr_result = self.ocr.ocr(image_array, cls=False)
2025-08-05 07:08:07 +00:00
# Extract text
2025-09-09 16:43:46 +00:00
ocr_text = ""
if ocr_result and ocr_result[0]:
text = [
line[1][0] if line and len(line) >= 2 and line[1] else ""
for line in ocr_result[0]
]
text = [t.strip() for t in text if t]
ocr_text = " ".join(text)
logger.info(f"OCR extracted {len(ocr_text)} characters")
return ocr_text
2025-08-05 07:08:07 +00:00
except Exception as e:
logger.error(f"OCR recognition error: {str(e)}")
return ""
2025-08-05 07:08:07 +00:00
class NanonetsOCRBackend(OCRBackend):
"""Nanonets OCR backend implementation using OpenAI API format"""
def __init__(self):
2025-08-05 07:08:07 +00:00
"""Initialize Nanonets OCR backend
2025-08-05 07:08:07 +00:00
Args:
api_key: API key for OpenAI API
base_url: Base URL for OpenAI API
model: Model name
"""
base_url = os.getenv("OCR_API_BASE_URL", "http://localhost:8000/v1")
api_key = os.getenv("OCR_API_KEY", "123")
timeout = 30
self.client = OpenAI(api_key=api_key, base_url=base_url, timeout=timeout)
self.model = os.getenv("OCR_MODEL", "nanonets/Nanonets-OCR-s")
logger.info(f"Nanonets OCR engine initialized with model: {self.model}")
self.temperature = 0.0
self.max_tokens = 15000
self.prompt = """## 任务说明
2025-08-05 07:08:07 +00:00
请从上传的文档中提取文字内容,严格按自然阅读顺序(从上到下,从左到右)输出,并遵循以下格式规范。
### 1. **文本处理**
* 按正常阅读顺序提取文字,语句流畅自然。
### 2. **表格**
* 所有表格统一转换为 **Markdown 表格格式**。
* 内容保持清晰、对齐整齐,便于阅读。
### 3. **公式**
* 所有公式转换为 **LaTeX 格式**,使用 `$$公式$$` 包裹。
### 4. **图片**
* 忽略图片信息
### 5. **链接**
* 不要猜测或补全不确定的链接地址。
"""
2025-08-05 07:08:07 +00:00
def predict(self, image: Union[str, bytes, Image.Image]) -> str:
"""Extract text from an image using Nanonets OCR
2025-08-05 07:08:07 +00:00
Args:
image: Image file path, bytes, or PIL Image object
2025-08-05 07:08:07 +00:00
Returns:
Extracted text
"""
if self.client is None:
logger.error("Nanonets OCR client not initialized")
return ""
2025-08-05 07:08:07 +00:00
try:
# Encode image to base64
img_base64 = endecode.decode_image(image)
2025-08-05 07:08:07 +00:00
if not img_base64:
return ""
2025-08-05 07:08:07 +00:00
# Call Nanonets OCR API
logger.info(f"Calling Nanonets OCR API with model: {self.model}")
response = self.client.chat.completions.create(
model=self.model,
messages=[
{
"role": "user",
"content": [
{
"type": "image_url",
"image_url": {
"url": f"data:image/png;base64,{img_base64}"
},
2025-08-05 07:08:07 +00:00
},
{
"type": "text",
"text": self.prompt,
},
],
}
],
temperature=self.temperature,
max_tokens=self.max_tokens,
2025-08-05 07:08:07 +00:00
)
return response.choices[0].message.content or ""
2025-08-05 07:08:07 +00:00
except Exception as e:
logger.error(f"Nanonets OCR prediction error: {str(e)}")
return ""
2025-08-05 07:08:07 +00:00
class OCREngine:
"""OCR Engine factory class"""
_instance: Dict[str, OCRBackend] = {}
2025-08-05 07:08:07 +00:00
@classmethod
def get_instance(cls, backend_type: str) -> OCRBackend:
2025-08-05 07:08:07 +00:00
"""Get OCR engine instance
2025-08-05 07:08:07 +00:00
Args:
backend_type: OCR backend type, one of: "paddle", "nanonets"
**kwargs: Additional arguments for the backend
2025-08-05 07:08:07 +00:00
Returns:
OCR engine instance or None if initialization fails
"""
backend_type = backend_type.lower()
if cls._instance.get(backend_type):
return cls._instance[backend_type]
logger.info(f"Initializing OCR engine with backend: {backend_type}")
if backend_type == "paddle":
cls._instance[backend_type] = PaddleOCRBackend()
elif backend_type == "nanonets":
cls._instance[backend_type] = NanonetsOCRBackend()
else:
cls._instance[backend_type] = DummyOCRBackend()
return cls._instance[backend_type]