Update services/vision.py

services/vision.py

@@ -2,21 +2,17 @@ import cv2
 import os
 import time
 from deepface import DeepFace
-from core.config import DETECTOR, MODELS, THRESHOLDS, MIN_FACE_AREA
+from core.config import DETECTOR, MODELS, THRESHOLDS, MIN_FACE_AREA, REQUIRED_AVERAGE_CONFIDENCE
 
 def process_frame_synchronous(frame):
     """
-    The Hard-Logic Pipeline:
-    1. Detects all faces.
-    2. Filters by quality.
-    3. Identifies every face with a confidence score.
-    4. Returns a complete report BEFORE the next frame is allowed.
+    The Smart-Hard Pipeline:
+    Calculates weighted average across models and returns failure details.
     """
     report = []
     pipeline_status = {"detected": 0, "identified": 0}
     
     try:
-        # Phase 1: Harvester (Detection)
         raw_faces = DeepFace.extract_faces(img_path=frame, detector_backend=DETECTOR, enforce_detection=True, align=True)
         pipeline_status["detected"] = len(raw_faces)
         
@@ -24,17 +20,13 @@ def process_frame_synchronous(frame):
             region = face_obj.get('facial_area', {})
             w, h = region.get('w', 0), region.get('h', 0)
             
-            # Quality Gate
             if (w * h) < MIN_FACE_AREA:
                 continue
                 
             x, y = region.get('x', 0), region.get('y', 0)
             cropped_face = frame[y:y+h, x:x+w]
-            
-            if cropped_face.size == 0:
-                continue
+            if cropped_face.size == 0: continue
 
-            # Phase 2: Interrogator (Consensus Engine)
             predictions = {}
             confidences = {}
             
@@ -48,35 +40,37 @@ def process_frame_synchronous(frame):
                     best_match = df.iloc[0]
                     dist = best_match["distance"]
                     
-                    if dist < THRESHOLDS[model]:
-                        name = os.path.basename(best_match['identity']).split('.')[0]
-                        predictions[model] = name
-                        # Calculate mathematical confidence percentage
-                        # Distance 0 = 100%, Distance 1.0 = 0%
-                        conf = round((1.0 - dist) * 100, 1)
-                        confidences[model] = conf
-                    else:
-                        predictions[model] = None
+                    # Convert distance to confidence (0.0 distance = 100%)
+                    conf = round((1.0 - dist) * 100, 1)
+                    name = os.path.basename(best_match['identity']).split('.')[0]
+                    
+                    predictions[model] = name
+                    confidences[model] = conf
                 else:
-                    predictions[model] = None
+                    predictions[model] = "Unknown"
+                    confidences[model] = 0.0
 
-            # Consensus Check
-            names_found = list(predictions.values())
-            if None not in names_found and len(set(names_found)) == 1:
-                # Both models agree!
-                confirmed_name = names_found[0]
-                # Use the primary model (ArcFace) confidence for the report
-                score = confidences.get("ArcFace", 0)
-                report.append({"name": confirmed_name, "confidence": score})
+            # --- SMART CONSENSUS LOGIC ---
+            # 1. Check if they agree on the NAME
+            all_names = [n for n in predictions.values() if n != "Unknown"]
+            unique_names = set(all_names)
+            
+            avg_conf = sum(confidences.values()) / len(MODELS)
+            
+            if len(unique_names) == 1 and avg_conf >= REQUIRED_AVERAGE_CONFIDENCE:
+                confirmed_name = all_names[0]
+                report.append({"status": "match", "name": confirmed_name, "score": avg_conf})
                 pipeline_status["identified"] += 1
             else:
-                # Disagreement or unknown
-                report.append({"name": "Unknown", "confidence": 0})
+                # Failure Report: Tell the user WHO we thought it was and WHAT the score was
+                best_guess = "Unknown"
+                if unique_names: best_guess = list(unique_names)[0]
+                report.append({"status": "fail", "name": best_guess, "score": avg_conf})
 
         return report, pipeline_status
 
     except ValueError:
-        return [], pipeline_status # No faces found
+        return [], pipeline_status
     except Exception as e:
         print(f"🔴 Critical Pipeline Error: {e}")
         return [], pipeline_status