Update processing/segmentation/hair_segmentation.py

processing/segmentation/hair_segmentation.py

@@ -262,7 +262,7 @@ def segment(self, frame: np.ndarray) -> np.ndarray:
             # Combine masks
             combined_mask = cv2.bitwise_or(hair_mask_hsv, hair_mask_lab)
             
-            # Morphological operations
+            # Morphological operations (using OpenCV instead of skimage)
             kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (7, 7))
             combined_mask = cv2.morphologyEx(combined_mask, cv2.MORPH_CLOSE, kernel)
             combined_mask = cv2.morphologyEx(combined_mask, cv2.MORPH_OPEN, kernel)
@@ -472,14 +472,16 @@ def _try_segment_with_model(self, frame: np.ndarray, model_name: str) -> Tuple[O
             return None, model_name, error_msg
     
     def _calculate_confidence(self, mask: np.ndarray) -> float:
-        """Calculate mask confidence"""
+        """Calculate mask confidence using OpenCV instead of skimage"""
         # Edge sharpness
         edges = cv2.Canny((mask * 255).astype(np.uint8), 50, 150)
         edge_ratio = np.sum(edges > 0) / mask.size
         
-        # Mask smoothness
-        gradient = np.gradient(mask)
-        smoothness = 1.0 / (1.0 + np.std(gradient))
+        # Mask smoothness using OpenCV Sobel instead of skimage gradient
+        grad_x = cv2.Sobel(mask, cv2.CV_64F, 1, 0, ksize=3)
+        grad_y = cv2.Sobel(mask, cv2.CV_64F, 0, 1, ksize=3)
+        gradient_magnitude = np.sqrt(grad_x**2 + grad_y**2)
+        smoothness = 1.0 / (1.0 + np.std(gradient_magnitude))
         
         return min(edge_ratio * 0.3 + smoothness * 0.7, 1.0)