Update app.py

app.py

@@ -1,182 +1,226 @@
 import gradio as gr
-import cv2
-import numpy as np
-import os
+import matplotlib.pyplot as plt
 import pandas as pd
-from datetime import datetime
-
-# Store evaluation data
-feedback_data = []
-
-##############################################
-# Utility: Detect plate color and vehicle type
-##############################################
-
-def classify_vehicle_by_plate_color(roi):
-
-    hsv = cv2.cvtColor(roi, cv2.COLOR_BGR2HSV)
-
-    # Masks for colors
-    masks = {}
-
-    # White
-    masks["white"] = cv2.inRange(hsv, np.array([0, 0, 180]), np.array([180, 60, 255]))
-
-    # Yellow
-    masks["yellow"] = cv2.inRange(hsv, np.array([15, 80, 80]), np.array([40, 255, 255]))
-
-    # Green
-    masks["green"] = cv2.inRange(hsv, np.array([35, 50, 50]), np.array([85, 255, 255]))
-
-    # Red
-    masks["red1"] = cv2.inRange(hsv, np.array([0, 70, 50]), np.array([10, 255, 255]))
-    masks["red2"] = cv2.inRange(hsv, np.array([170, 70, 50]), np.array([180, 255, 255]))
-    masks["red"] = masks["red1"] + masks["red2"]
-
-    # Blue
-    masks["blue"] = cv2.inRange(hsv, np.array([90, 50, 50]), np.array([130, 255, 255]))
+import os
+import random
+from PIL import ImageDraw
+
+# ===============================
+# GLOBAL VARIABLES
+# ===============================
+total_vehicles = 0
+ev_count = 0
+total_co2_saved = 0
+
+DISTANCE_KM = 10
+CO2_PER_KM_PETROL = 0.150
+CO2_SAVED_PER_EV = DISTANCE_KM * CO2_PER_KM_PETROL
+
+FEEDBACK_FILE = "feedback.csv"
+
+if not os.path.exists(FEEDBACK_FILE):
+    pd.DataFrame(columns=["Predicted_Label", "Feedback"]).to_csv(FEEDBACK_FILE, index=False)
+
+# ===============================
+# SIMULATED NUMBER PLATE COLOUR DETECTION
+# Replace with real color detection later
+# ===============================
+def detect_plate_colour():
+    return random.choice(["White", "Yellow", "Green", "Black"])
+
+def get_vehicle_type_from_colour(colour):
+    mapping = {
+        "White": "Private Vehicle",
+        "Yellow": "Commercial Vehicle",
+        "Green": "Electric Vehicle",
+        "Black": "Rental Vehicle"
+    }
+    return mapping.get(colour, "Unknown")
+
+# ===============================
+# DASHBOARD
+# ===============================
+def generate_dashboard():
+    global total_vehicles, ev_count
+    non_ev = total_vehicles - ev_count
+
+    fig, ax = plt.subplots()
+    ax.bar(["EV", "Non-EV"], [ev_count, non_ev])
+    ax.set_title("Vehicle Distribution")
+    ax.set_ylabel("Count")
+    return fig
+
+# ===============================
+# EVALUATION SUMMARY
+# ===============================
+def get_evaluation_summary():
+    df = pd.read_csv(FEEDBACK_FILE)
+    total = len(df)
+    correct = len(df[df["Feedback"] == "Correct"])
+    incorrect = len(df[df["Feedback"] == "Incorrect"])
 
-    # Count pixels
-    color_counts = {color: np.sum(mask) for color, mask in masks.items()}
+    if total == 0:
+        return "Evaluation Summary:\nNo feedback yet."
 
-    dominant_color = max(color_counts, key=color_counts.get)
+    precision = correct / total
 
-    # Classification logic
-    if dominant_color == "white":
-        return "Private Vehicle"
-    elif dominant_color == "yellow":
-        return "Commercial Vehicle"
-    elif dominant_color == "green":
-        return "Electric Vehicle (EV)"
-    elif dominant_color == "red":
-        return "Temporary Registration Vehicle"
-    elif dominant_color == "blue":
-        return "Diplomatic Vehicle"
-    else:
-        return "Unknown Vehicle Type"
+    return f"""Evaluation Summary
+Total: {total}
+Correct: {correct}
+Incorrect: {incorrect}
+Accuracy: {precision:.2f}
+"""
 
-##############################################
-# Main Detection Function
-##############################################
+# ===============================
+# DETECTION FUNCTION
+# ===============================
+def detect_image(image):
 
-def detect_vehicles(image):
+    global total_vehicles, ev_count, total_co2_saved
 
     if image is None:
-        return None, "No image uploaded."
-
-    img = cv2.cvtColor(np.array(image), cv2.COLOR_RGB2BGR)
-    gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
-
-    # Simple plate-like contour detection
-    edges = cv2.Canny(gray, 100, 200)
-    contours, _ = cv2.findContours(edges, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
-
-    detected_info = []
-    count = 0
-
-    for cnt in contours:
-        x, y, w, h = cv2.boundingRect(cnt)
-
-        # Plate aspect ratio filter
-        if 2 < w / h < 6 and w > 100:
-            plate_roi = img[y:y+h, x:x+w]
-            vehicle_type = classify_vehicle_by_plate_color(plate_roi)
+        return None, "Upload image first.", "", "", "", "", "", ""
 
-            cv2.rectangle(img, (x, y), (x+w, y+h), (0, 255, 0), 2)
-            cv2.putText(img, vehicle_type,
-                        (x, y-10),
-                        cv2.FONT_HERSHEY_SIMPLEX,
-                        0.6,
-                        (0, 255, 0),
-                        2)
+    plate_colour = detect_plate_colour()
+    vehicle_type = get_vehicle_type_from_colour(plate_colour)
+    plate_number = "KA01AB1234"
 
-            detected_info.append(vehicle_type)
-            count += 1
+    total_vehicles += 1
+    co2_saved_this = 0
 
-    if count == 0:
-        summary = "No vehicles detected."
-    else:
-        summary = f"{count} Vehicle(s) Detected: " + ", ".join(detected_info)
+    if vehicle_type == "Electric Vehicle":
+        ev_count += 1
+        co2_saved_this = CO2_SAVED_PER_EV
+        total_co2_saved += co2_saved_this
 
-    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
+    ev_percent = (ev_count / total_vehicles) * 100
 
-    return img, summary
+    # Draw boxes
+    img = image.copy()
+    draw = ImageDraw.Draw(img)
+    w, h = img.size
 
+    vehicle_box = [w*0.1, h*0.2, w*0.9, h*0.8]
+    plate_box = [w*0.4, h*0.6, w*0.7, h*0.75]
 
-##############################################
-# Feedback System
-##############################################
+    draw.rectangle(vehicle_box, outline="red", width=4)
+    draw.text((vehicle_box[0], vehicle_box[1]-20),
+              f"{vehicle_type}", fill="red")
 
-def submit_feedback(is_correct):
-    global feedback_data
+    draw.rectangle(plate_box, outline="green", width=4)
+    draw.text((plate_box[0], plate_box[1]-20),
+              f"{plate_number} ({plate_colour})", fill="green")
 
-    feedback_data.append({
-        "timestamp": datetime.now(),
-        "correct": is_correct
-    })
+    fig, ax = plt.subplots()
+    ax.imshow(img)
+    ax.axis("off")
 
-    return generate_summary()
-
-
-def generate_summary():
-    total = len(feedback_data)
-    if total == 0:
-        return "No evaluations yet."
-
-    correct = sum(1 for f in feedback_data if f["correct"])
-    accuracy = (correct / total) * 100
-
-    return f"""
-📊 Evaluation Summary  
-Total Samples: {total}  
-Correct: {correct}  
-Accuracy: {accuracy:.2f}%
+    result_text = f"""
+Vehicle Type: {vehicle_type}
+Plate Colour: {plate_colour}
+Plate Number: {plate_number}
+CO₂ Saved: {co2_saved_this:.2f} kg
 """
 
-
-##############################################
-# Gradio UI
-##############################################
-
+    total_card = f"### 🚘 Total: {total_vehicles}"
+    ev_card = f"### ⚡ EV: {ev_count}"
+    percent_card = f"### 📊 EV Rate: {ev_percent:.2f}%"
+    co2_card = f"### 🌱 CO₂ Saved: {total_co2_saved:.2f} kg"
+
+    dashboard_fig = generate_dashboard()
+
+    summary = get_evaluation_summary()
+
+    return fig, result_text, total_card, ev_card, percent_card, co2_card, dashboard_fig, vehicle_type, summary
+
+# ===============================
+# FEEDBACK SAVE
+# ===============================
+def save_feedback(predicted_label, feedback):
+    df = pd.read_csv(FEEDBACK_FILE)
+    df = pd.concat([df, pd.DataFrame(
+        [{"Predicted_Label": predicted_label, "Feedback": feedback}]
+    )], ignore_index=True)
+    df.to_csv(FEEDBACK_FILE, index=False)
+    return "Saved!", get_evaluation_summary()
+
+# ===============================
+# RESET DATABASE
+# ===============================
+def reset_database():
+    pd.DataFrame(columns=["Predicted_Label", "Feedback"]).to_csv(FEEDBACK_FILE, index=False)
+    return "Database Reset!", "Evaluation Summary:\nNo feedback yet."
+
+# ===============================
+# UI
+# ===============================
 with gr.Blocks() as demo:
 
-    gr.Markdown("## Smart Traffic Vehicle Classification System")
+    gr.Markdown("## 🚦 Smart Vehicle Detection Based on Number Plate Colour")
 
     with gr.Row():
+        img_input = gr.Image(type="pil", label="Upload Vehicle Image")
+        img_output = gr.Plot()
 
-        with gr.Column(scale=2):
+    detect_btn = gr.Button("Detect", size="sm")
 
-            image_input = gr.Image(type="pil", label="Upload Vehicle Image")
+    result_box = gr.Textbox(label="Detection Result", lines=5)
 
-            detect_btn = gr.Button("Detect", size="sm")
+    with gr.Row():
+        total_card = gr.Markdown("### 🚘 Total: 0")
+        ev_card = gr.Markdown("### ⚡ EV: 0")
+        percent_card = gr.Markdown("### 📊 EV Rate: 0%")
+        co2_card = gr.Markdown("### 🌱 CO₂ Saved: 0 kg")
 
-            output_image = gr.Image(label="Output Image")
-            output_text = gr.Textbox(label="Detection Summary")
+    dashboard_plot = gr.Plot()
 
-        with gr.Column(scale=1):
+    predicted_label_state = gr.State()
 
-            gr.Markdown("### Feedback")
+    detect_btn.click(
+        fn=detect_image,
+        inputs=[img_input],
+        outputs=[
+            img_output,
+            result_box,
+            total_card,
+            ev_card,
+            percent_card,
+            co2_card,
+            dashboard_plot,
+            predicted_label_state,
+            gr.Markdown()  # evaluation summary placeholder
+        ]
+    )
 
-            correct_btn = gr.Button("Correct", size="sm")
-            incorrect_btn = gr.Button("Incorrect", size="sm")
+    # -------------------------
+    # FEEDBACK + EVALUATION ROW
+    # -------------------------
+    with gr.Row():
 
-            summary_box = gr.Textbox(label="Evaluation Summary")
+        with gr.Column(scale=1):
+            correct_btn = gr.Button("✔", size="sm")
+            incorrect_btn = gr.Button("✘", size="sm")
+            feedback_status = gr.Textbox(label="Feedback", lines=1)
 
-    # Button actions
-    detect_btn.click(
-        fn=detect_vehicles,
-        inputs=image_input,
-        outputs=[output_image, output_text]
-    )
+        with gr.Column(scale=1):
+            evaluation_summary = gr.Textbox(label="Evaluation Summary", lines=6)
 
     correct_btn.click(
-        fn=lambda: submit_feedback(True),
-        outputs=summary_box
+        fn=save_feedback,
+        inputs=[predicted_label_state, gr.State("Correct")],
+        outputs=[feedback_status, evaluation_summary]
     )
 
     incorrect_btn.click(
-        fn=lambda: submit_feedback(False),
-        outputs=summary_box
+        fn=save_feedback,
+        inputs=[predicted_label_state, gr.State("Incorrect")],
+        outputs=[feedback_status, evaluation_summary]
+    )
+
+    reset_btn = gr.Button("Reset Database", size="sm")
+    reset_btn.click(
+        fn=reset_database,
+        outputs=[feedback_status, evaluation_summary]
     )
 
 demo.launch(theme=gr.themes.Soft())