解码base64图片后ocr图片成功

机器狗后台服务/app/api/v1/api.py

@@ -4,14 +4,20 @@ from sqlalchemy.ext.asyncio import AsyncSession
 from app.core.database import get_db
 from app.services.imageServices import ImageService
 from app.schemas.image import ImageBase
+from app.schemas.ocr import OCRRequest, OCRResponse
 from app.util.responseHttp import ResponseUtil
 from app.util.baiduOCR import BaiduOCR
+import base64
+import io
+from PIL import Image
+import os
+import uuid
 # from app.crud.event import event
 # from app.schemas.event import EventList, EventDetail, EventUpdate, EventQuery, TestEvent
 
 baiduOCR = BaiduOCR()
 
-router = APIRouter()
+router = APIRouter(prefix="/api/v1", tags=["ocr"])
 
 @router.get("/hello")
 async def get_hello():
@@ -45,3 +51,41 @@ async def test_ocr(
     # print(result)
     return ResponseUtil.success(msg="success", data=result)
     # return ResponseUtil.success(msg="success")
+
+
+@router.post("/ocr_from_base64")
+async def ocr_from_base64(request: OCRRequest):
+    """
+    从base64图片数据进行OCR识别
+    """
+    try:
+        # 移除base64数据的前缀（如果有）
+        image_base64 = request.image_base64
+        if image_base64.startswith('data:image'):
+            # 格式如: data:image/jpeg;base64,/9j/4AAQSkZJRgABAQAAAQ...
+            image_base64 = image_base64.split(',')[1]
+        
+        # 解码base64数据
+        image_data = base64.b64decode(image_base64)
+        
+        # 将字节数据转换为PIL Image对象
+        image = Image.open(io.BytesIO(image_data))
+        
+        # 创建临时文件路径
+        temp_dir = "./tmp/ocr_images"
+        os.makedirs(temp_dir, exist_ok=True)
+        temp_filename = f"{uuid.uuid4()}.{request.image_type or 'jpg'}"
+        temp_path = os.path.join(temp_dir, temp_filename)
+        
+        # 保存图片到临时文件
+        image.save(temp_path)
+        
+        # 使用PaddleOCR进行识别
+        result = baiduOCR.ocr(temp_path)
+        
+        # 删除临时文件
+        os.remove(temp_path)
+        
+        return ResponseUtil.success(msg="OCR识别成功", data=result[0])
+    except Exception as e:
+        return ResponseUtil.error(msg=f"OCR识别失败: {str(e)}", data=None)

机器狗后台服务/app/schemas/ocr.py

@@ -0,0 +1,11 @@
+from pydantic import BaseModel
+from typing import Optional
+
+class OCRRequest(BaseModel):
+    image_base64: str
+    image_type: Optional[str] = "jpg"  # 图片类型，如 jpg, png 等
+    
+class OCRResponse(BaseModel):
+    success: bool
+    message: str
+    data: Optional[list] = None

机器狗后台服务/test_ocr_base64.py

@@ -0,0 +1,54 @@
+import base64
+import requests
+import json
+
+def image_to_base64(image_path):
+    """
+    将图片文件转换为base64编码
+    """
+    with open(image_path, "rb") as image_file:
+        encoded_string = base64.b64encode(image_file.read()).decode('utf-8')
+    return encoded_string
+
+def test_ocr_api(image_path, api_url="http://10.0.0.202:12342/api/v1/ocr_from_base64"):
+    """
+    测试OCR API接口
+    """
+    # 将图片转换为base64
+    image_base64 = image_to_base64(image_path)
+    
+    with open("base64.txt", "w") as f:
+        f.write(repr(image_base64))
+    
+    # 准备请求数据
+    payload = {
+        "image_base64": image_base64,
+        "image_type": "jpg"  # 根据实际图片类型修改
+    }
+    
+    # 发送POST请求
+    headers = {
+        "Content-Type": "application/json"
+    }
+    
+    try:
+        response = requests.post(api_url, data=json.dumps(payload), headers=headers)
+        if response.status_code == 200:
+            result = response.json()
+            print("OCR识别结果:")
+            print(json.dumps(result, ensure_ascii=False, indent=2))
+            return result
+        else:
+            print(f"请求失败，状态码: {response.status_code}")
+            print(response.text)
+            return None
+    except Exception as e:
+        print(f"请求异常: {str(e)}")
+        return None
+
+if __name__ == "__main__":
+    # 测试图片路径，请根据实际情况修改
+    test_image_path = "/home/admin-root/haotian/康达瑞贝斯机器狗/data_image/001读表图片/2c7cc83019e7388a7041101da92c9829_frame_000000.jpg"
+    
+    # 调用测试函数
+    test_ocr_api(test_image_path)