更新OCR版本，Bata版，还不能上线

util/image_util.py

@@ -1,9 +1,12 @@
 import logging
 import math
 import urllib.request
+from io import BytesIO
 
 import cv2
 import numpy
+from PIL import Image
+from PIL.ExifTags import TAGS
 from paddleclas import PaddleClas
 from tenacity import retry, stop_after_attempt, wait_random
 
@@ -14,20 +17,36 @@ def read(image_path):
     """
     从网络或本地读取图片
     :param image_path: 网络或本地路径
-    :return: NumPy数组形式的图片
+    :return: NumPy数组形式的图片, EXIF数据
     """
     if image_path.startswith("http"):
         # 发送HTTP请求并获取图像数据
         resp = urllib.request.urlopen(image_path, timeout=60)
         # 将数据读取为字节流
         image_data = resp.read()
-        # 将字节流转换为NumPy数组
-        image_np = numpy.frombuffer(image_data, numpy.uint8)
-        # 解码NumPy数组为OpenCV图像格式
-        image = cv2.imdecode(image_np, cv2.IMREAD_COLOR)
     else:
-        image = cv2.imread(image_path)
-    return image
+        with open(image_path, "rb") as f:
+            image_data = f.read()
+
+    # 解析EXIF信息（基于原始字节流）
+    exif_data = {}
+    try:
+        # 用PIL打开原始字节流
+        with Image.open(BytesIO(image_data)) as img:
+            # 获取EXIF字典
+            exif_info = img._getexif()
+            if exif_info:
+                # 将EXIF标签的数字ID转换为可读名称（如36867对应"DateTimeOriginal"）
+                for tag_id, value in exif_info.items():
+                    tag_name = TAGS.get(tag_id, tag_id)
+                    exif_data[tag_name] = value
+    except Exception as e:
+        logging.error("解析EXIF信息失败", exc_info=e)
+    # 将字节流转换为NumPy数组
+    image_np = numpy.frombuffer(image_data, numpy.uint8)
+    # 解码NumPy数组为OpenCV图像格式
+    image = cv2.imdecode(image_np, cv2.IMREAD_COLOR)
+    return image, exif_data
 
 
 def capture(image, rectangle):
@@ -61,7 +80,7 @@ def split(image, ratio=1.414, overlap=0.05, x_compensation=3):
     """
     split_result = []
     if isinstance(image, str):
-        image = read(image)
+        image, _ = read(image)
     height, width = image.shape[:2]
     hw_ratio = height / width
     wh_ratio = width / height