Update app.py

app.py

@@ -2,45 +2,52 @@ import streamlit as st
 import cv2
 import numpy as np
 import tensorflow as tf
-from tensorflow.keras.applications.mobilenet import preprocess_input
-from tensorflow.keras.applications.mobilenet import decode_predictions
+from tensorflow.keras.applications.mobilenet import preprocess_input, decode_predictions
 
 st.title("物体识别应用")
 st.write("通过摄像头识别物体，从左到右显示主要物体的名称")
 
-# 设置摄像头
-video_capture = cv2.VideoCapture(0)
-stframe = st.empty()
-
-# 加载 MobileNet SSD 预训练模型
+# 加载 MobileNet 预训练模型
 model = tf.keras.applications.MobileNet(weights="imagenet")
 
+# 打开摄像头函数
+def open_camera():
+    video_capture = cv2.VideoCapture(0)
+    return video_capture
+
+# 物体识别函数
 def detect_objects(frame):
-    # 预处理图像
     image_resized = cv2.resize(frame, (224, 224))
     image_array = np.expand_dims(image_resized, axis=0)
     processed_image = preprocess_input(image_array)
     
-    # 使用模型进行预测
     preds = model.predict(processed_image)
-    decoded_preds = decode_predictions(preds, top=3)[0]  # 取前3个结果
+    decoded_preds = decode_predictions(preds, top=3)[0]
     objects = [f"{label}: {round(score * 100, 2)}%" for (_, label, score) in decoded_preds]
     return objects
 
-# 读取摄像头流并显示
-while True:
-    ret, frame = video_capture.read()
-    if not ret:
-        st.write("无法读取摄像头数据。")
-        break
-
-    # 检测物体
-    objects = detect_objects(frame)
-    detected_text = " | ".join(objects)
-
-    # 显示检测结果
-    cv2.putText(frame, detected_text, (10, 30), cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 255, 0), 2, cv2.LINE_AA)
+# 按钮控制摄像头启动
+if st.button("开始摄像头"):
+    video_capture = open_camera()
+    stframe = st.empty()
     
-    # 将 OpenCV 图像格式转换为 Streamlit 显示格式
-    frame_rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
-    stframe.image(frame_rgb, caption="检测到的物体", channels="RGB")
+    while video_capture.isOpened():
+        ret, frame = video_capture.read()
+        if not ret:
+            st.write("无法读取摄像头数据。")
+            break
+        
+        objects = detect_objects(frame)
+        detected_text = " | ".join(objects)
+        
+        # 显示检测结果在帧上
+        cv2.putText(frame, detected_text, (10, 30), cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 255, 0), 2, cv2.LINE_AA)
+        
+        # 转换并显示在 Streamlit
+        frame_rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
+        stframe.image(frame_rgb, caption="检测到的物体", channels="RGB")
+
+    video_capture.release()
+    cv2.destroyAllWindows()
+else:
+    st.write("点击上方按钮以启动摄像头")