Update app.py

app.py

@@ -15,16 +15,16 @@ model.to('cuda' if torch.cuda.is_available() else 'cpu')  # Use GPU if available
 # Constants for LBW decision and video processing
 STUMPS_WIDTH = 0.2286  # meters (width of stumps)
 BALL_DIAMETER = 0.073  # meters (approx. cricket ball diameter)
-FRAME_RATE = 20  # Input video frame rate (to be updated dynamically)
+FRAME_RATE = 20  # Default frame rate, updated dynamically
 SLOW_MOTION_FACTOR = 3  # For very slow motion (3x slower)
-CONF_THRESHOLD = 0.4  # Increased confidence threshold for better detection
-IMPACT_ZONE_Y = 0.8  # Adjusted fraction of frame height for impact zone
+CONF_THRESHOLD = 0.25  # Lowered further to improve detection chances
+IMPACT_ZONE_Y = 0.8  # Fraction of frame height for impact zone
 IMPACT_VELOCITY_THRESHOLD = 1000  # Pixels/second for detecting impact
 PITCH_LENGTH = 20.12  # meters (standard cricket pitch length)
 STUMPS_HEIGHT = 0.71  # meters (stumps height)
 CAMERA_HEIGHT = 2.0  # meters (assumed camera height)
 CAMERA_DISTANCE = 10.0  # meters (assumed camera distance from pitch)
-MAX_POSITION_JUMP = 50  # Increased for smoother trajectory
+MAX_POSITION_JUMP = 50  # For smoother trajectory filtering
 
 def process_video(video_path):
     if not os.path.exists(video_path):
@@ -34,6 +34,10 @@ def process_video(video_path):
     frame_width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
     frame_height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
     FRAME_RATE = cap.get(cv2.CAP_PROP_FPS) or 20  # Use actual frame rate or default
+    # Adjust image size to be multiple of 32 for YOLO
+    stride = 32
+    img_width = ((frame_width + stride - 1) // stride) * stride
+    img_height = ((frame_height + stride - 1) // stride) * stride
     frames = []
     ball_positions = []
     detection_frames = []
@@ -46,15 +50,21 @@ def process_video(video_path):
             break
         frame_count += 1
         frames.append(frame.copy())
-        # Enhance frame contrast for better detection
-        frame = cv2.convertScaleAbs(frame, alpha=1.2, beta=10)
-        results = model.predict(frame, conf=CONF_THRESHOLD, imgsz=(frame_height, frame_width), iou=0.5, max_det=1)
-        detections = 0
-        for detection in results[0].boxes:
-            if detection.cls == 0:  # Class 0 is the ball
-                detections += 1
-                if detections == 1:  # Only consider frames with exactly one detection
+        # Enhance frame contrast and sharpness for better detection
+        frame = cv2.convertScaleAbs(frame, alpha=1.5, beta=20)
+        kernel = np.array([[-1, -1, -1], [-1, 9, -1], [-1, -1, -1]])
+        frame = cv2.filter2D(frame, -1, kernel)
+        results = model.predict(frame, conf=CONF_THRESHOLD, imgsz=(img_height, img_width), iou=0.5, max_det=3)
+        detections = sum(1 for detection in results[0].boxes if detection.cls == 0)
+        if detections == 1:  # Only process frames with exactly one ball detection
+            for detection in results[0].boxes:
+                if detection.cls == 0:  # Class 0 is the ball
                     x1, y1, x2, y2 = detection.xyxy[0].cpu().numpy()
+                    # Scale coordinates back to original frame size
+                    x1 = x1 * frame_width / img_width
+                    x2 = x2 * frame_width / img_width
+                    y1 = y1 * frame_height / img_height
+                    y2 = y2 * frame_height / img_height
                     ball_positions.append([(x1 + x2) / 2, (y1 + y2) / 2])
                     detection_frames.append(frame_count - 1)
                     cv2.rectangle(frame, (int(x1), int(y1)), (int(x2), int(y2)), (0, 255, 0), 2)
@@ -65,9 +75,9 @@ def process_video(video_path):
     cap.release()
 
     if not ball_positions:
-        debug_log.append("No balls detected in any frame")
+        debug_log.append("No frames with exactly one ball detection")
     else:
-        debug_log.append(f"Total ball detections: {len(ball_positions)}")
+        debug_log.append(f"Total frames with one ball detection: {len(ball_positions)}")
         debug_log.append(f"Video resolution: {frame_width}x{frame_height}")
         debug_log.append(f"Video frame rate: {FRAME_RATE}")
 
@@ -84,7 +94,7 @@ def pixel_to_3d(x, y, frame_height, frame_width):
 
 def estimate_trajectory(ball_positions, frames, detection_frames):
     if len(ball_positions) < 2:
-        return None, None, None, None, None, None, None, None, None, "Error: Fewer than 2 ball detections for trajectory"
+        return None, None, None, None, None, None, None, None, None, "Error: Fewer than 2 frames with one ball detection"
     frame_height, frame_width = frames[0].shape[:2]
 
     # Filter out sudden changes in position for continuous trajectory
@@ -92,7 +102,7 @@ def estimate_trajectory(ball_positions, frames, detection_frames):
     filtered_frames = [detection_frames[0]]
     for i in range(1, len(ball_positions)):
         prev_pos = filtered_positions[-1]
-        curr_pos = box_positions[i]
+        curr_pos = ball_positions[i]
         distance = np.sqrt((curr_pos[0] - prev_pos[0])**2 + (curr_pos[1] - prev_pos[1])**2)
         if distance <= MAX_POSITION_JUMP:
             filtered_positions.append(curr_pos)