Update app.py

app.py

@@ -38,7 +38,6 @@ gps_coordinates: List[List[float]] = []
 last_metrics: Dict[str, Any] = {}
 frame_count: int = 0
 SAVE_IMAGE_INTERVAL = 1
-DETECTION_CLASSES = ["Longitudinal", "Pothole", "Transverse", "Toll gate"]  # Updated for Toll gates
 MAX_IMAGES = 500
 
 # Model setup
@@ -63,7 +62,7 @@ model_colors = {
     'Toll gates': (0, 0, 255)  # Blue
 }
 
-# Helper functions (unchanged)
+# Helper functions
 def zip_all_outputs(report_path: str, video_path: str, chart_path: str, map_path: str) -> str:
     zip_path = os.path.join(OUTPUT_DIR, f"drone_analysis_outputs_{datetime.now().strftime('%Y%m%d_%H%M%S')}.zip")
     try:
@@ -370,7 +369,7 @@ def generate_report(
         log_entries.append(f"Error: Failed to save report: {str(e)}")
         return ""
 
-def process_video(video, selected_model, resize_width=1920, resize_height=1080, frame_skip=10):
+def process_video(video, selected_model, resize_width=1920, resize_height=1080, frame_skip=1):
     global frame_count, last_metrics, detected_counts, detected_issues, gps_coordinates, log_entries
     frame_count = 0
     detected_counts.clear()
@@ -436,15 +435,16 @@ def process_video(video, selected_model, resize_width=1920, resize_height=1080,
         frame = cv2.resize(frame, (out_width, out_height))
         resize_times.append((time.time() - frame_start) * 1000)
 
-        if not check_image_quality(frame, input_resolution):
-            continue
+        # Comment out quality check to process all frames
+        # if not check_image_quality(frame, input_resolution):
+        #     continue
 
         annotated_frame = frame.copy()
         frame_detections = []
         inference_start = time.time()
 
         for model_name, model in use_models.items():
-            results = model(annotated_frame, verbose=False, conf=0.5, iou=0.7)
+            results = model(annotated_frame, verbose=False, conf=0.25, iou=0.45)
             for result in results:
                 for box in result.boxes:
                     x1, y1, x2, y2 = map(int, box.xyxy[0].tolist())
@@ -454,18 +454,17 @@ def process_video(video, selected_model, resize_width=1920, resize_height=1080,
                     cv2.rectangle(annotated_frame, (x1, y1), (x2, y2), color, 2)
                     cv2.putText(annotated_frame, label, (x1, y1 - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.8, color, 2)
 
-                    if model.names[class_id] in DETECTION_CLASSES:
-                        detection_data = {
-                            "label": model.names[class_id],
-                            "model": model_name,
-                            "box": [x1, y1, x2, y2],
-                            "conf": float(box.conf[0]),
-                            "gps": [17.385044 + (frame_count * 0.0001), 78.486671 + (frame_count * 0.0001)],
-                            "timestamp": f"{int(frame_count / fps // 60):02d}:{int(frame_count / fps % 60):02d}",
-                            "frame": frame_count,
-                            "path": os.path.join(CAPTURED_FRAMES_DIR, f"detected_{frame_count:06d}.jpg")
-                        }
-                        frame_detections.append(detection_data)
+                    detection_data = {
+                        "label": model.names[class_id],
+                        "model": model_name,
+                        "box": [x1, y1, x2, y2],
+                        "conf": float(box.conf[0]),
+                        "gps": [17.385044 + (frame_count * 0.0001), 78.486671 + (frame_count * 0.0001)],
+                        "timestamp": f"{int(frame_count / fps // 60):02d}:{int(frame_count / fps % 60):02d}",
+                        "frame": frame_count,
+                        "path": os.path.join(CAPTURED_FRAMES_DIR, f"detected_{frame_count:06d}.jpg")
+                    }
+                    frame_detections.append(detection_data)
 
         inference_times.append((time.time() - inference_start) * 1000)
 
@@ -580,7 +579,7 @@ with gr.Blocks(theme=gr.themes.Soft(primary_hue="orange")) as iface:
             )
             width_slider = gr.Slider(320, 1920, value=1920, label="Output Width", step=1)
             height_slider = gr.Slider(240, 1080, value=1080, label="Output Height", step=1)
-            skip_slider = gr.Slider(1, 20, value=10, label="Frame Skip", step=1)
+            skip_slider = gr.Slider(1, 20, value=1, label="Frame Skip", step=1)
             process_btn = gr.Button("Process Video", variant="primary")
         with gr.Column(scale=1):
             metrics_output = gr.Textbox(label="Detection Metrics", lines=5, interactive=False)