Update app.py

app.py

@@ -1,67 +1,40 @@
 import gradio as gr
-import cv2
-import numpy as np
-from video_utils import stream_video
-from usecases.crack_detector import detect_cracks
-# from usecases.pothole_detector import detect_potholes  # Enable for potholes
-
-video_path = "assets/Drone_Video.mp4"
-
-def process_frame(frame):
-    crack_frame, crack_results = detect_cracks(frame)
-    # pothole_frame, pothole_results = detect_potholes(frame)
-    
-    # Combine annotated outputs (can be composited visually if needed)
-    combined = crack_frame  # For now just cracks
-
-    cracks = []
-    for box in crack_results.boxes:
-        cls = crack_results.names[int(box.cls[0])]
-        conf = round(float(box.conf[0]), 2)
-        cracks.append([cls, conf])
-
-    return combined, len(crack_results.boxes), cracks
-
-with gr.Blocks(title="Drone Road Inspection Dashboard") as demo:
-    gr.Markdown("## 🚁 Drone Road Inspection Dashboard")
-    gr.Markdown("**Status:** 🟢 Running")
-
-    with gr.Row():
-        live_feed = gr.Image(label="📺 Live Drone Feed", interactive=False)
-        crack_metrics = gr.Label(label="📊 Crack Metrics")
-
-    with gr.Row():
-        live_logs = gr.Textbox(label="🧾 Live Logs", interactive=False, lines=6)
-        crack_trend = gr.Plot(label="📈 Crack Trend")
-        crack_severity = gr.Label(label="🚨 Crack Severity")
-
-    with gr.Row():
-        crack_map = gr.Image(label="🗺️ Crack Locations Map", interactive=False)
-        crack_list = gr.Dataframe(headers=["Type", "Confidence"], label="Detected Cracks (Last 100+)")
-
-    with gr.Row():
-        pause = gr.Button("⏸ Pause")
-        resume = gr.Button("▶️ Resume")
-        interval_slider = gr.Slider(minimum=0.1, maximum=5.0, value=0.5, step=0.1, label="Frame Interval (seconds)")
-
-    def run_analysis(interval):
-        logs = []
-        crack_data = []
-
-        for frame in stream_video(video_path, interval):
-            processed_frame, crack_count, cracks = process_frame(frame)
-            logs.append(f"Frame processed: {crack_count} cracks detected.")
-            crack_data += cracks[-100:]
-
-            yield {
-                live_feed: processed_frame,
-                crack_metrics: {"Total Cracks": crack_count},
-                live_logs: "\n".join(logs[-10:]),
-                crack_list: crack_data
-            }
-
-    demo.load(run_analysis, inputs=[interval_slider], outputs=[
-        live_feed, crack_metrics, live_logs, crack_list
-    ])
+from utils.frame_extractor import extract_frames
+from utils.image_utils import save_frame
+from services import under_construction, operations_maintenance, road_safety, plantation
+import os
+
+def process_video(video, category):
+    video_path = video.name
+    frames = extract_frames(video_path)
+    results = []
+    os.makedirs('outputs', exist_ok=True)
+
+    for idx, frame in enumerate(frames[:5]):  # Limiting to 5 frames for demonstration
+        if category == "Under Construction":
+            detection = under_construction.detect_under_construction(frame)
+        elif category == "Operations and Maintenance":
+            detection = operations_maintenance.detect_operations_maintenance(frame)
+        elif category == "Road Safety":
+            detection = road_safety.detect_road_safety(frame)
+        elif category == "Plantation":
+            detection = plantation.detect_plantation(frame)
+        else:
+            detection = None
+
+        output_path = f"outputs/frame_{idx}.jpg"
+        save_frame(frame, output_path)
+        results.append(output_path)
+
+    return results
+
+with gr.Blocks() as demo:
+    gr.Markdown("# 🚧 Drone Surveillance Analysis")
+    video_input = gr.Video(label="Upload Drone Footage")
+    category = gr.Dropdown(choices=["Under Construction", "Operations and Maintenance", "Road Safety", "Plantation"], label="Select Category")
+    output_gallery = gr.Gallery(label="Detected Frames").style(grid=[2], height="auto")
+    analyze_button = gr.Button("Analyze")
+
+    analyze_button.click(fn=process_video, inputs=[video_input, category], outputs=output_gallery)
 
 demo.launch()