Merge branch 'main' of https://github.com/sustech-cs304/team-project-25spring-9

Author: mumuzi2023

Process_Image/.coverage

@@ -1031,18 +1031,18 @@
   {
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2025-05-25T05:53:25.542254Z",
-     "start_time": "2025-05-25T05:53:22.499891Z"
+     "end_time": "2025-05-25T16:21:02.485162Z",
+     "start_time": "2025-05-25T16:20:54.661065Z"
     }
    },
    "cell_type": "code",
    "source": [
     "import requests\n",
     "# url = \"http://localhost:8123/generate_capetion/\"\n",
-    "url = \"http://localhost:8123/auto_tag/\"\n",
+    "# url = \"http://localhost:8123/auto_tag/\"\n",
     "# url = \"http://localhost:8123/extract_exif/\"\n",
     "# url = \"http://localhost:8123/face_recognition/\"\n",
-    "# url = \"http://localhost:8123/process_image/\"\n",
+    "url = \"http://localhost:8123/process_image/\"\n",
     "files = {'file': open('photos/25.jpg', 'rb')}\n",
     "response = requests.post(url, files=files)\n",
     "print(response.json())"
@@ -1053,28 +1053,28 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "{'tags': ['computer', 'monitor']}\n"
+      "{'Timestamp': '2025-05-25 01:01:41', 'Latitude': None, 'Longitude': None, 'Address': None, 'Camera/Device': 'Xiaomi 23013RK75C', 'Caption': 'a computer monitor', 'AutoTags': ['monitor', 'computer'], 'PersonLabel': []}\n"
      ]
     }
    ],
-   "execution_count": 63
+   "execution_count": 11
   },
   {
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2025-05-25T08:05:36.461104Z",
-     "start_time": "2025-05-25T08:05:19.547419Z"
+     "end_time": "2025-05-25T13:42:42.232445Z",
+     "start_time": "2025-05-25T13:42:33.238079Z"
     }
    },
    "cell_type": "code",
    "source": [
     "import requests\n",
     "\n",
     "# 替换为你本地或线上 FastAPI 服务的地址\n",
-    "url = \"http://localhost:8123/style_transfer/\"\n",
+    "url = \"http://10.16.60.67:8123/style_transfer/\"\n",
     "\n",
     "# 要使用的风格编号\n",
-    "style_index = 0  # 可以修改为你想测试的风格编号\n",
+    "style_index = 1  # 可以修改为你想测试的风格编号\n",
     "\n",
     "# 打开图片并构造 multipart/form-data 请求\n",
     "# with open(image_path, \"rb\") as image_file:\n",
@@ -1104,11 +1104,11 @@
      "output_type": "stream",
      "text": [
       "风格化任务提交成功：\n",
-      "{'task_id': '46733e90-ef4d-4a9b-b5dc-15082cd3cd88', 'result_url': 'https://dashscope-result-wlcb.oss-cn-wulanchabu.aliyuncs.com/1d/06/20250525/81750fc9/20250525160537191180_style0_q0laxient5.jpg?Expires=1748246743&OSSAccessKeyId=LTAI5tQZd8AEcZX6KZV4G8qL&Signature=nwrlcZ6tRjivk5D7VD%2FJwI5wxyg%3D'}\n"
+      "{'task_id': 'bababc39-a67b-495a-bb96-ed5ef6e57f73', 'result_url': 'https://dashscope-result-wlcb.oss-cn-wulanchabu.aliyuncs.com/1d/be/20250525/81750fc9/20250525214243526580_style0_20hetuzcym.jpg?Expires=1748266969&OSSAccessKeyId=LTAI5tQZd8AEcZX6KZV4G8qL&Signature=13B1BEq0ty1UAUsF0vDUQ73rHZ8%3D'}\n"
      ]
     }
    ],
-   "execution_count": 7
+   "execution_count": 10
   },
   {
    "metadata": {

Process_Image/API.ipynb

@@ -396,8 +396,24 @@ def generate_video(image_paths: list[str]) -> str:
 
     for idx, img_path in enumerate(image_paths):
         im = Image.open(img_path)
+        # max_size = (1920, 1080)
+        # im.thumbnail(max_size, Image.Resampling.LANCZOS)
         max_size = (1920, 1080)
+        min_size = (1920, 1080)  # 建议设为较合理的最低值
+
+        # 先缩小到不超过 max_size，保持比例
         im.thumbnail(max_size, Image.Resampling.LANCZOS)
+
+        # 如果太小，再放大（保持比例）
+        width, height = im.size
+        if width < min_size[0] or height < min_size[1]:
+            scale_w = min_size[0] / width
+            scale_h = min_size[1] / height
+            scale = max(scale_w, scale_h)  # 保证放大后两个维度都 ≥ 最小尺寸
+            new_width = int(width * scale)
+            new_height = int(height * scale)
+            im = im.resize((new_width, new_height), Image.Resampling.LANCZOS)
+
         effect_type = random.choice([0, 1])
         if effect_type == 0:
             x_speed = (im.width - im.width * 0.8) / duration

Process_Image/process_image.py

@@ -0,0 +1,42 @@
+import os
+from fastapi.testclient import TestClient
+from process_image import app
+
+client = TestClient(app)
+
+SAMPLE_IMAGE = os.path.join(os.path.dirname(__file__), "sample.jpg")
+
+
+def test_extract_exif():
+    with open(SAMPLE_IMAGE, "rb") as f:
+        response = client.post("/extract_exif/", files={"file": ("sample.jpg", f, "image/jpeg")})
+    assert response.status_code == 200
+    data = response.json()
+    assert "Timestamp" in data
+    assert "Latitude" in data
+    assert "Longitude" in data
+
+
+def test_generate_caption():
+    with open(SAMPLE_IMAGE, "rb") as f:
+        response = client.post("/generate_caption/", files={"file": ("sample.jpg", f, "image/jpeg")})
+    assert response.status_code == 200
+    assert "caption" in response.json()
+
+
+def test_auto_tag():
+    with open(SAMPLE_IMAGE, "rb") as f:
+        response = client.post("/auto_tag/", files={"file": ("sample.jpg", f, "image/jpeg")})
+    assert response.status_code == 200
+    assert "tags" in response.json()
+    assert isinstance(response.json()["tags"], list)
+
+
+def test_process_image():
+    with open(SAMPLE_IMAGE, "rb") as f:
+        response = client.post("/process_image", files={"file": ("sample.jpg", f, "image/jpeg")})
+    assert response.status_code == 200
+    data = response.json()
+    assert "Caption" in data
+    assert "AutoTags" in data
+    assert "PersonLabel" in data

Process_Image/tests/__pycache__/test_api.cpython-312-pytest-7.4.4.pyc

@@ -10,6 +10,10 @@
 
 ![img.png](backend.assets/img.png)
 
+[🖼️ 图像智能处理后端 API - Swagger UI](http://10.16.60.67:8123/docs)
+
+![image-20250525221802032](backend/image-20250525221802032.png)
+
 ## Tests for backend
 
 Technology: jcoco