Update app.py

app.py

@@ -1,68 +1,54 @@
+# Install necessary libraries
 import streamlit as st
+from huggingface_hub import hf_hub_download
 from ultralytics import YOLO
-from PIL import Image, ImageDraw
-import requests
-import tempfile
-import os
-
-# Download YOLOv8 model from Hugging Face
-model_url = "https://huggingface.co/krishnamishra8848/Nepal_Vehicle_License_Plates_Detection_Version2/resolve/main/best.pt"
-model_path = "best.pt"
-
-# Download the model if not already downloaded
-if not os.path.exists(model_path):
-    st.info("Downloading model from Hugging Face...")
-    with open(model_path, "wb") as f:
-        response = requests.get(model_url)
-        f.write(response.content)
-    st.success("Model downloaded successfully!")
-
-# Load the YOLOv8 model
-model = YOLO(model_path)
-
-# App title
+import cv2
+import numpy as np
+from PIL import Image
+
+# Step 1: Download the YOLO model weights from your Hugging Face repository
+weights_path = hf_hub_download(repo_id="krishnamishra8848/Nepal-Vehicle-License-Plate-Detection", filename="best.pt")
+
+# Step 2: Load the YOLO model
+model = YOLO(weights_path)
+
+# Step 3: Function to process and display results
+def detect_license_plate(image):
+    # Convert the PIL image to a numpy array
+    img = np.array(image)
+    img = cv2.cvtColor(img, cv2.COLOR_RGB2BGR)
+
+    # Perform inference
+    results = model(img)
+
+    # Draw bounding boxes and confidence scores
+    for result in results:
+        if hasattr(result, 'boxes') and result.boxes is not None:
+            for box, conf in zip(result.boxes.xyxy, result.boxes.conf):
+                x1, y1, x2, y2 = map(int, box)  # Convert to integers
+                cv2.rectangle(img, (x1, y1), (x2, y2), (0, 255, 0), 2)  # Green rectangle
+                label = f"{conf:.2f}"
+                cv2.putText(img, label, (x1, y1 - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 0), 2)
+
+    # Convert back to RGB for Streamlit display
+    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
+    return Image.fromarray(img)
+
+# Streamlit Interface
 st.title("Nepal Vehicle License Plate Detection")
-st.write("Upload an image to detect vehicle license plates along with their confidence scores.")
-
-# Upload image
-uploaded_image = st.file_uploader("Choose an image", type=["jpg", "jpeg", "png"])
+st.write("Upload an image to detect license plates.")
 
-if uploaded_image is not None:
-    # Save the uploaded image to a temporary file
-    temp_file = tempfile.NamedTemporaryFile(delete=False, suffix=".jpg")  # Ensure proper file extension
-    temp_file.write(uploaded_image.read())
-    temp_file.close()
+# File uploader
+uploaded_file = st.file_uploader("Choose an image...", type=["jpg", "jpeg", "png"])
 
+if uploaded_file is not None:
     # Display the uploaded image
-    image = Image.open(temp_file.name)
+    image = Image.open(uploaded_file)
     st.image(image, caption="Uploaded Image", use_column_width=True)
 
-    # Run YOLOv8 inference
-    with st.spinner("Running detection..."):
-        results = model(temp_file.name)
-
-    # Draw bounding boxes and confidence scores on the image
-    draw = ImageDraw.Draw(image)
-    results_table = []
-    for box in results[0].boxes:
-        # Get bounding box coordinates and confidence score
-        x_min, y_min, x_max, y_max = map(int, box.xyxy[0].tolist())
-        confidence = box.conf.item()
-        label = f"Plate: {confidence:.2f}"
-
-        # Draw rectangle and label
-        draw.rectangle([(x_min, y_min), (x_max, y_max)], outline="red", width=3)
-        draw.text((x_min, y_min - 10), label, fill="red")
-
-        # Append detection to the table
-        results_table.append({"x_min": x_min, "y_min": y_min, "x_max": x_max, "y_max": y_max, "confidence": confidence})
-
-    # Display the resulting image with bounding boxes
-    st.image(image, caption="Detected Image", use_column_width=True)
-
-    # Show individual detections in a table
-    st.write("### Detection Results")
-    st.write(results_table)
+    # Run detection
+    st.write("Processing...")
+    result_image = detect_license_plate(image)
 
-# Remove temporary files (optional cleanup)
-os.unlink(temp_file.name)
+    # Display the results
+    st.image(result_image, caption="Detection Results", use_column_width=True)