Update app.py

app.py

@@ -2,82 +2,35 @@ import gradio as gr
 import cv2
 import numpy as np
 from ultralytics import YOLO
-from pathlib import Path
-import tempfile
-import os
+from importlib.metadata import version
 
-# 初始化模型占位符
+# 动态组件选择
+GRADIO_VERSION = version("gradio")
+if int(GRADIO_VERSION.split('.')[0]) >= 4 and int(GRADIO_VERSION.split('.')[1]) >= 12:
+    Webcam = gr.Webcam
+else:
+    class Webcam(gr.Image):
+        def __init__(self, ​**kwargs):
+            kwargs.update(source="webcam", streaming=True)
+            super().__init__(**kwargs)
+
+# YOLO模型初始化
 model = None
 
-def load_custom_model(model_file):
+def load_model(model_file):
     global model
-    try:
-        # 保存上传文件
-        save_path = f"/tmp/{model_file.name}"
-        with open(save_path, "wb") as f:
-            f.write(model_file.read())
-        
-        # 加载模型
-        model = YOLO(save_path)
-        return "✅ 模型加载成功！"
-    except Exception as e:
-        return f"❌ 加载失败：{str(e)}"
-
-def process_frame(frame):
-    if model is None:
-        raise gr.Error("请先上传模型文件")
-    
-    results = model.predict(
-        source=frame,
-        device="cpu",
-        conf=0.5,
-        verbose=False
-    )
-    return cv2.cvtColor(results[0].plot(), cv2.COLOR_BGR2RGB)
+    model = YOLO(model_file.name)
+    return "✅ 模型加载成功"
 
-# Gradio界面（兼容旧版）
-with gr.Blocks(title="YOLO检测系统") as demo:
-    gr.Markdown("# 🔍 自定义YOLO检测系统")
-    
-    with gr.Tab("🔧 模型管理"):
-        upload_btn = gr.UploadButton(
-            "上传.pt模型文件",
-            file_types=[".pt"],
-            variant="primary"
-        )
-        status = gr.Textbox(label="状态")
-        upload_btn.upload(load_custom_model, upload_btn, status)
-    
+with gr.Blocks() as demo:
     with gr.Tab("📷 实时检测"):
-        webcam = gr.Image(
-            source="webcam",
-            streaming=True,
-            label="摄像头画面"
-        )
-        output = gr.Image(label="检测结果")
+        webcam = Webcam(label="摄像头")
+        output = gr.Image()
         webcam.stream(
-            fn=lambda x: process_frame(x),
+            fn=lambda x: model.predict(x)[0].plot(),
             inputs=webcam,
             outputs=output
         )
-    
-    with gr.Tab("🖼️ 图片检测"):
-        img_input = gr.Image(type="filepath")
-        img_output = gr.Image()
-        img_input.change(
-            fn=lambda x: process_frame(cv2.imread(x)),
-            inputs=img_input,
-            outputs=img_output
-        )
-    
-    with gr.Tab("🎥 视频检测"):
-        vid_input = gr.Video()
-        vid_output = gr.Video()
-        vid_input.change(
-            fn=lambda x: video_pipeline(x),
-            inputs=vid_input,
-            outputs=vid_output
-        )
 
 if __name__ == "__main__":
-    demo.launch(server_name="0.0.0.0")
+    demo.launch()