Update src/streamlit_app.py

src/streamlit_app.py

@@ -1,8 +1,4 @@
-from fastai.vision.all import *
-from io import BytesIO
-import requests
-import streamlit as st
-
+from PIL import Image
 import numpy as np
 import torch
 import time
@@ -10,151 +6,101 @@ import cv2
 from numpy import random
 import os
 import sys
+from io import BytesIO
+import requests
+import streamlit as st
 
-# 加入上層目錄到模組搜尋路徑中
-sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__), '..')))
+# 將 yolov7 的副程式路徑加入系統中
+sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__), 'yolov7')))
 
 from models.experimental import attempt_load
-from utils.general import check_img_size, check_requirements, check_imshow, non_max_suppression, apply_classifier, \
-    scale_coords, xyxy2xywh, strip_optimizer, set_logging, increment_path
+from utils.general import check_img_size, non_max_suppression, scale_coords
 from utils.plots import plot_one_box
 
-def letterbox(img, new_shape=(640, 640), color=(114, 114, 114), auto=True, scaleFill=False, scaleup=True, stride=32):
-    # Resize and pad image while meeting stride-multiple constraints
-    shape = img.shape[:2]  # current shape [height, width]
+# Device 設定
+device = torch.device("cpu")
+
+# 圖像 resize（YOLO 專用）
+def letterbox(img, new_shape=(640, 640), color=(114, 114, 114), auto=False, scaleFill=False, scaleup=True, stride=32):
+    shape = img.shape[:2]
     if isinstance(new_shape, int):
         new_shape = (new_shape, new_shape)
-
-    # Scale ratio (new / old)
     r = min(new_shape[0] / shape[0], new_shape[1] / shape[1])
-    if not scaleup:  # only scale down, do not scale up (for better test mAP)
+    if not scaleup:
         r = min(r, 1.0)
-
-    # Compute padding
-    ratio = r, r  # width, height ratios
+    ratio = r, r
     new_unpad = int(round(shape[1] * r)), int(round(shape[0] * r))
-    dw, dh = new_shape[1] - new_unpad[0], new_shape[0] - new_unpad[1]  # wh padding
-    if auto:  # minimum rectangle
-        dw, dh = np.mod(dw, stride), np.mod(dh, stride)  # wh padding
-    elif scaleFill:  # stretch
-        dw, dh = 0.0, 0.0
-        new_unpad = (new_shape[1], new_shape[0])
-        ratio = new_shape[1] / shape[1], new_shape[0] / shape[0]  # width, height ratios
-
-    dw /= 2  # divide padding into 2 sides
+    dw, dh = new_shape[1] - new_unpad[0], new_shape[0] - new_unpad[1]
+    dw /= 2
     dh /= 2
-
-    if shape[::-1] != new_unpad:  # resize
-        img = cv2.resize(img, new_unpad, interpolation=cv2.INTER_LINEAR)
+    img = cv2.resize(img, new_unpad, interpolation=cv2.INTER_LINEAR)
     top, bottom = int(round(dh - 0.1)), int(round(dh + 0.1))
     left, right = int(round(dw - 0.1)), int(round(dw + 0.1))
-    img = cv2.copyMakeBorder(img, top, bottom, left, right, cv2.BORDER_CONSTANT, value=color)  # add border
+    img = cv2.copyMakeBorder(img, top, bottom, left, right, cv2.BORDER_CONSTANT, value=color)
     return img, ratio, (dw, dh)
+# 模型路徑（確保 best.pt 放在同一層）
+weight_path = os.path.join(os.path.dirname(__file__), "best.pt")
 
-def detect_modify(img0, model, conf=0.4, imgsz=640, conf_thres = 0.25, iou_thres=0.45):
-    st.image(img0, caption="Your image", use_column_width=True)
+# 載入模型
+ckpt = torch.load(weight_path, map_location=device)
+model = ckpt['ema' if ckpt.get('ema') else 'model'].float().fuse().eval()
 
-    stride = int(model.stride.max())  # model stride
-    imgsz = check_img_size(imgsz, s=stride)  # check img_size
+# 加入自定義類別名稱（必要，否則無法畫框標籤）
+model.names = ['WithMask', 'WithoutMask']  # 替換為你的類別
 
-    # Padded resize
-    img0 = cv2.cvtColor(np.asarray(img0), cv2.COLOR_RGB2BGR)
-    img = letterbox(img0, imgsz, stride=stride)[0]
-    # Convert
-    img = img[:, :, ::-1].transpose(2, 0, 1)  # BGR to RGB, to 3x416x416
+# 推論主函式
+def detect(img_pil, conf_thres=0.25, iou_thres=0.45, imgsz=640):
+    img0 = np.array(img_pil.convert('RGB'))
+    img = cv2.cvtColor(img0, cv2.COLOR_RGB2BGR)
+    img, _, _ = letterbox(img, new_shape=imgsz)
+    img = img[:, :, ::-1].transpose(2, 0, 1)  # BGR to RGB, CHW
     img = np.ascontiguousarray(img)
-
-    
-    # Get names and colors
-    names = model.module.names if hasattr(model, 'module') else model.names
-    colors = [[random.randint(0, 255) for _ in range(3)] for _ in names]
-
-    # Run inference
-    old_img_w = old_img_h = imgsz
-    old_img_b = 1
-
-    t0 = time.time()
-    img = torch.from_numpy(img).to(device)
-    # img /= 255.0  # 0 - 255 to 0.0 - 1.0
-    img = img/255.0
+    img = torch.from_numpy(img).to(device).float() / 255.0
     if img.ndimension() == 3:
         img = img.unsqueeze(0)
 
-    # Inference
-    # t1 = time_synchronized()
-    with torch.no_grad():   # Calculating gradients would cause a GPU memory leak
+    with torch.no_grad():
         pred = model(img)[0]
-    # t2 = time_synchronized()
-
-    # Apply NMS
-    pred = non_max_suppression(pred, conf_thres, iou_thres)
-    # t3 = time_synchronized()
-
-    # Process detections
-    # for i, det in enumerate(pred):  # detections per image
-        
-    gn = torch.tensor(img0.shape)[[1, 0, 1, 0]]  # normalization gain whwh
+        pred = non_max_suppression(pred, conf_thres, iou_thres)
 
     det = pred[0]
-    if len(det):
-        # Rescale boxes from img_size to im0 size
-        det[:, :4] = scale_coords(img.shape[2:], det[:, :4], img0.shape).round()
-
-        # Print results
-        s = ''
-        for c in det[:, -1].unique():
-            n = (det[:, -1] == c).sum()  # detections per class
-            s += f"{n} {names[int(c)]}{'s' * (n > 1)}, "  # add to string
+    names = model.names
+    colors = [[random.randint(0, 255) for _ in range(3)] for _ in names]
 
-        # Write results
-        for *xyxy, conf, cls in reversed(det):
+    if det is not None and len(det):
+        det[:, :4] = scale_coords(img.shape[2:], det[:, :4], img0.shape).round()
+        for *xyxy, conf, cls in det:
             label = f'{names[int(cls)]} {conf:.2f}'
-            plot_one_box(xyxy, img0, label=label, color=colors[int(cls)], line_thickness=1)
+            plot_one_box(xyxy, img0, label=label, color=colors[int(cls)], line_thickness=2)
 
-    f"""
-    ### Prediction result:
-    """
-    img0 = cv2.cvtColor(np.asarray(img0), cv2.COLOR_BGR2RGB)
-    st.image(img0, caption="Prediction Result", use_column_width=True)
+    return Image.fromarray(img0)
+st.set_page_config(page_title="YOLOv7 Mask Detection", layout="centered")
+st.title("🛡️ YOLOv7 Mask Detection")
+st.write("請上傳圖片或提供圖片網址，辨識是否有戴口罩。")
 
-#set paramters
+option = st.radio("選擇輸入方式：", ["上傳圖片", "輸入圖片網址"])
 
-# 取得目前檔案 (streamlit_app.py) 所在的目錄
-current_dir = os.path.dirname(os.path.abspath(__file__))
+img = None
 
-# 回到根目錄後組合出 .pkl 檔案的路徑
-weight_path = os.path.join(current_dir, 'best.pt')
-
-imgsz = 640
-conf = 0.4
-conf_thres = 0.25
-iou_thres=0.45
-device = torch.device("cpu")
-path = "./"
-
-# Load model
-#model = attempt_load(weight_path, map_location=torch.device('cpu'))  # load FP32 model
-ckpt = torch.load(weight_path, map_location=torch.device('cpu'), weights_only=False)
-model = ckpt['ema' if ckpt.get('ema') else 'model'].float().fuse().eval()
+if option == "上傳圖片":
+    uploaded_file = st.file_uploader("請上傳圖片（格式如 JPG, PNG）", type=["jpg", "jpeg", "png"])
+    if uploaded_file:
+        try:
+            img = Image.open(uploaded_file).convert("RGB")
+        except:
+            st.error("⚠️ 無法讀取圖片，請確認格式是否正確。")
 
-"""
-# YOLOv7
-Detect whether a person is wearing a face mask or not, using image or URL input.
-"""
-option = st.radio("", ["Upload Image", "Image URL"])
-
-if option == "Upload Image":
-    uploaded_file = st.file_uploader("Please upload an image.")
-
-    if uploaded_file is not None:
-        img = PILImage.create(uploaded_file)
-        detect_modify(img, model, conf=conf, imgsz=imgsz, conf_thres=conf_thres, iou_thres=iou_thres)
-else:
-    url = st.text_input("Please input a url.")
-    if url != "":
+elif option == "輸入圖片網址":
+    url = st.text_input("請貼上圖片網址")
+    if url:
         try:
             response = requests.get(url)
-            pil_img = PILImage.create(BytesIO(response.content))
-            detect_modify(pil_img, model, conf=conf, imgsz=imgsz, conf_thres=conf_thres, iou_thres=iou_thres)
+            img = Image.open(BytesIO(response.content)).convert("RGB")
         except:
-            st.text("Problem reading image from", url)
+            st.error("⚠️ 無法從網址讀取圖片，請確認連結是否正確。")
+
+if img:
+    st.image(img, caption="原始圖片", use_column_width=True)
+    with st.spinner("模型推論中，請稍候..."):
+        result_img = detect(img)
+    st.image(result_img, caption="模型辨識結果", use_column_width=True)