86 lines
3.4 KiB
Python
86 lines
3.4 KiB
Python
|
import base64
|
|||
|
|
import json
|
|||
|
|
import logging
|
|||
|
|
import requests
|
|||
|
|
from Config.Config import GLM_API_KEY, GLM_MODEL_NAME, GLM_BASE_URL
|
|||
|
|
|
|||
|
|
# 获取模块专属日志器
|
|||
|
|
logger = logging.getLogger(__name__)
|
|||
|
|
|
|||
|
|
|
|||
|
|
def extract_text_from_image(image_path):
|
|||
|
|
"""
|
|||
|
|
从图片中提取文字内容（排除几何图形中的文字）
|
|||
|
|
Args:
|
|||
|
|
image_path: 图片文件的绝对路径
|
|||
|
|
Returns:
|
|||
|
|
str: 提取的文字内容，若失败则返回空字符串
|
|||
|
|
"""
|
|||
|
|
try:
|
|||
|
|
# 读取图片并转换为base64
|
|||
|
|
with open(image_path, "rb") as image_file:
|
|||
|
|
base64_image = base64.b64encode(image_file.read()).decode('utf-8')
|
|||
|
|
image_url = f"data:image/png;base64,{base64_image}"
|
|||
|
|
|
|||
|
|
# 构建API请求
|
|||
|
|
# 合并重复的要求字段，使用清晰的编号系统
|
|||
|
|
prompt = "【任务】提取图像中的数学题文字内容（排除几何图形），用【】包裹结果。"
|
|||
|
|
prompt += "【专业术语】需准确识别：将军饮马、动点、等量代换、两点之间线段最短、垂线段最短"
|
|||
|
|
prompt += "【格式要求】1. 术语错误需修正；2. 保证句子语法通顺；3. 不添加解释；4. 不包含思考过程"
|
|||
|
|
url = GLM_BASE_URL
|
|||
|
|
headers = {
|
|||
|
|
"Authorization": f"Bearer {GLM_API_KEY}",
|
|||
|
|
"Content-Type": "application/json"
|
|||
|
|
}
|
|||
|
|
data = {
|
|||
|
|
"model": GLM_MODEL_NAME,
|
|||
|
|
"messages": [{
|
|||
|
|
"role": "user",
|
|||
|
|
"content": [
|
|||
|
|
{"type": "text", "text": prompt},
|
|||
|
|
{"type": "image_url", "image_url": {"url": image_url}}
|
|||
|
|
]
|
|||
|
|
}],
|
|||
|
|
"stream": True
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
# 发送请求并处理流式响应
|
|||
|
|
extracted_text = []
|
|||
|
|
seen_content = set() # 用于去重的集合
|
|||
|
|
with requests.post(url, headers=headers, json=data, stream=True) as response:
|
|||
|
|
for chunk in response.iter_lines():
|
|||
|
|
if chunk:
|
|||
|
|
decoded = chunk.decode('utf-8')
|
|||
|
|
if decoded.startswith('[DONE]'):
|
|||
|
|
logger.info("文字提取完成")
|
|||
|
|
break
|
|||
|
|
try:
|
|||
|
|
# 解析流式响应内容
|
|||
|
|
json_data = json.loads(decoded[5:]) # 移除""前缀
|
|||
|
|
content = json_data["choices"][0]["delta"].get('content')
|
|||
|
|
if content and len(content) > 0 and content not in seen_content:
|
|||
|
|
seen_content.add(content)
|
|||
|
|
extracted_text.append(content)
|
|||
|
|
except Exception as e:
|
|||
|
|
logger.warning(f"解析响应块失败: {str(e)}")
|
|||
|
|
|
|||
|
|
# 拼接结果并返回
|
|||
|
|
return ''.join(extracted_text)
|
|||
|
|
|
|||
|
|
except FileNotFoundError:
|
|||
|
|
logger.error(f"图片文件不存在: {image_path}")
|
|||
|
|
except Exception as e:
|
|||
|
|
logger.error(f"文字提取失败: {str(e)}", exc_info=True)
|
|||
|
|
return ""
|
|||
|
|
|
|||
|
|
|
|||
|
|
# 示例用法
|
|||
|
|
if __name__ == "__main__":
|
|||
|
|
# 配置日志（实际项目中应在主程序统一配置）
|
|||
|
|
logging.basicConfig(level=logging.INFO)
|
|||
|
|
# 测试图片路径
|
|||
|
|
test_image = r"D:\dsWork\dsProject\dsLightRag\Test\extracted\a62dce9d67c818accf94113aabefe172\1_1_TXT.png"
|
|||
|
|
# 调用提取函数
|
|||
|
|
text = extract_text_from_image(test_image)
|
|||
|
|
print("提取结果:\n", text)
|