File size: 6,188 Bytes
48922fa
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
"""
Image analysis engine for processing and analyzing images.
"""
from typing import Dict, Any, List, Optional
import io
from PIL import Image
import torch
from torchvision import transforms
from transformers import AutoFeatureExtractor, AutoModelForImageClassification
import face_recognition
import numpy as np
from tenacity import retry, stop_after_attempt, wait_exponential

class ImageEngine:
    def __init__(self):
        # Initialize image classification model
        self.feature_extractor = AutoFeatureExtractor.from_pretrained(
            "microsoft/resnet-50"
        )
        self.model = AutoModelForImageClassification.from_pretrained(
            "microsoft/resnet-50"
        )
        
        # Set up image transforms
        self.transform = transforms.Compose([
            transforms.Resize(256),
            transforms.CenterCrop(224),
            transforms.ToTensor(),
            transforms.Normalize(
                mean=[0.485, 0.456, 0.406],
                std=[0.229, 0.224, 0.225]
            )
        ])
    
    @retry(stop=stop_after_attempt(3), wait=wait_exponential(multiplier=1, min=4, max=10))
    async def analyze_image(self, image_data: bytes) -> Dict[str, Any]:
        """Analyze image content and detect objects/faces."""
        try:
            # Load image
            image = Image.open(io.BytesIO(image_data)).convert('RGB')
            
            # Prepare image for model
            inputs = self.feature_extractor(images=image, return_tensors="pt")
            
            # Get model predictions
            with torch.no_grad():
                outputs = self.model(**inputs)
                probs = outputs.logits.softmax(-1)
                
            # Get top predictions
            top_probs, top_indices = torch.topk(probs, k=5)
            
            # Convert predictions to list
            predictions = [
                {
                    "label": self.model.config.id2label[idx.item()],
                    "confidence": prob.item()
                }
                for prob, idx in zip(top_probs[0], top_indices[0])
            ]
            
            # Analyze faces
            np_image = np.array(image)
            face_locations = face_recognition.face_locations(np_image)
            face_encodings = face_recognition.face_encodings(np_image, face_locations)
            
            faces = []
            for i, (face_encoding, face_location) in enumerate(zip(face_encodings, face_locations)):
                face = {
                    "id": i + 1,
                    "location": {
                        "top": face_location[0],
                        "right": face_location[1],
                        "bottom": face_location[2],
                        "left": face_location[3]
                    },
                    "encoding": face_encoding.tolist()
                }
                faces.append(face)
            
            # Get image metadata
            metadata = {
                "format": image.format,
                "mode": image.mode,
                "size": image.size,
                "width": image.width,
                "height": image.height
            }
            
            return {
                "predictions": predictions,
                "faces": faces,
                "metadata": metadata
            }
            
        except Exception as e:
            return {"error": str(e)}
    
    async def compare_faces(self, face1_data: bytes, face2_data: bytes) -> Dict[str, Any]:
        """Compare two faces and determine if they are the same person."""
        try:
            # Load and process first image
            image1 = face_recognition.load_image_file(io.BytesIO(face1_data))
            face1_encoding = face_recognition.face_encodings(image1)
            
            if not face1_encoding:
                return {"error": "No face found in first image"}
            
            # Load and process second image
            image2 = face_recognition.load_image_file(io.BytesIO(face2_data))
            face2_encoding = face_recognition.face_encodings(image2)
            
            if not face2_encoding:
                return {"error": "No face found in second image"}
            
            # Compare faces
            results = face_recognition.compare_faces(
                [face1_encoding[0]], face2_encoding[0]
            )
            
            # Calculate face distance (lower means more similar)
            face_distance = face_recognition.face_distance(
                [face1_encoding[0]], face2_encoding[0]
            )
            
            return {
                "match": bool(results[0]),
                "confidence": float(1 - face_distance[0]),
                "distance": float(face_distance[0])
            }
            
        except Exception as e:
            return {"error": str(e)}
    
    async def search_similar_faces(self, 
                                 target_encoding: List[float],
                                 face_database: List[Dict[str, Any]],
                                 threshold: float = 0.6) -> List[Dict[str, Any]]:
        """Search for similar faces in a database of face encodings."""
        try:
            matches = []
            target_encoding = np.array(target_encoding)
            
            for face_data in face_database:
                if "encoding" not in face_data:
                    continue
                    
                current_encoding = np.array(face_data["encoding"])
                distance = face_recognition.face_distance([target_encoding], current_encoding)[0]
                
                if distance < threshold:
                    matches.append({
                        "face_id": face_data.get("id"),
                        "confidence": float(1 - distance),
                        "metadata": face_data.get("metadata", {})
                    })
            
            # Sort matches by confidence
            matches.sort(key=lambda x: x["confidence"], reverse=True)
            
            return matches
            
        except Exception as e:
            return [{"error": str(e)}]