Détection du Groupe Sanguin à partir d'une Image - API/Website Python

Here is the complete code for a blood grouping detection project using Python, Flask, and OpenCV. It includes image preprocessing, feature extraction, classification (based on agglutination detection), and both a REST API and a simple web interface.

Entire code in one file `app.py`, plus `templates/index.html` for the web UI.

---

### 1. Project Structure

``` blood_group_detector/ │ ├── app.py # Main Flask application (API + web) ├── templates/ │ └── index.html # Simple web interface ├── requirements.txt # Dependencies └── uploads/ # Folder to store uploaded images (auto-created) ```

---

### 2. `requirements.txt`

``` flask==2.3.3 opencv-python==4.8.1 numpy==1.24.4 pillow==10.1.0 ```

---

### 3. `app.py` (Complete Backend)

```python import os import cv2 import numpy as np from flask import Flask, request, jsonify, render_template, redirect, url_for from werkzeug.utils import secure_filename import uuid

app = Flask(__name__) app.config['UPLOAD_FOLDER'] = 'uploads' app.config['ALLOWED_EXTENSIONS'] = {'png', 'jpg', 'jpeg'}

# Ensure upload folder exists os.makedirs(app.config['UPLOAD_FOLDER'], exist_ok=True)

# Blood group mapping based on agglutination pattern # Order: [Anti-A, Anti-B, Anti-D (Rh)] # 0 = no agglutination, 1 = agglutination BLOOD_GROUP_MAP = { (1, 0, 0): 'A+', (1, 0, 1): 'A-', (0, 1, 0): 'B+', (0, 1, 1): 'B-', (1, 1, 0): 'AB+', (1, 1, 1): 'AB-', (0, 0, 0): 'O+', (0, 0, 1): 'O-' }

def allowed_file(filename): return '.' in filename and filename.rsplit('.', 1)[1].lower() in app.config['ALLOWED_EXTENSIONS']

def preprocess_image(image_path): """ Step 1: Image preprocessing - Load, resize, convert to grayscale, apply Gaussian blur, thresholding """ img = cv2.imread(image_path) if img is None: raise ValueError("Could not read image") # Resize to standard size for consistency img = cv2.resize(img, (600, 400)) # Convert to grayscale gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY) # Apply Gaussian blur to reduce noise blurred = cv2.GaussianBlur(gray, (5, 5), 0) # Adaptive thresholding to highlight agglutination regions thresh = cv2.adaptiveThreshold(blurred, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C, cv2.THRESH_BINARY_INV, 11, 2) return img, thresh

def extract_features(thresh_img): """ Step 2: Feature extraction - Detect circular regions (agglutination clusters) using contour detection - Split image into 3 vertical regions (for Anti-A, Anti-B, Anti-D) """ height, width = thresh_img.shape region_width = width // 3 # Split into 3 zones zones = [ thresh_img[:, 0:region_width], # Zone 1: Anti-A thresh_img[:, region_width:2*region_width], # Zone 2: Anti-B thresh_img[:, 2*region_width:3*region_width] # Zone 3: Anti-D ] features = [] for zone in zones: # Find contours contours, _ = cv2.findContours(zone, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE) # Count significant circular contours (agglutinated cells appear as small circles) significant_count = 0 for cnt in contours: area = cv2.contourArea(cnt) if area > 50: # Filter out small noise # Check if contour is approximately circular perimeter = cv2.arcLength(cnt, True) if perimeter > 0: circularity = 4 * np.pi * area / (perimeter * perimeter) if circularity > 0.4: # Circle-like shape significant_count += 1 # Also compute density of white pixels as additional feature density = np.sum(zone > 0) / (zone.shape[0] * zone.shape[1]) # Combine features: count and density features.append((significant_count, density)) return features

def classify_blood_group(features): """ Step 3: Classification - Decide agglutination (clumping) based on features - If count is high or density high -> agglutination present """ results = [] for count, density in features: # Heuristic: agglutination detected if many significant contours or high pixel density if count > 3 or density > 0.05: results.append(1) # Agglutination present else: results.append(0) # No agglutination # Convert to tuple for mapping result_tuple = tuple(results) # Check if valid if result_tuple in BLOOD_GROUP_MAP: blood_group = BLOOD_GROUP_MAP[result_tuple] return blood_group, results else: return "Unknown", results

def detect_blood_group(image_path): """ Main pipeline: preprocess -> extract -> classify """ original, thresh = preprocess_image(image_path) features = extract_features(thresh) blood_group, agglutination = classify_blood_group(features) return { 'blood_group': blood_group, 'agglutination': { 'anti_A': 'Positive' if agglutination[0] else 'Negative', 'anti_B': 'Positive' if agglutination[1] else 'Negative', 'anti_D': 'Positive' if agglutination[2] else 'Negative' } }

# ------------------- API Endpoint ------------------- @app.route('/api/detect', methods=['POST']) def api_detect(): """ REST API endpoint: POST with file field 'image' -> returns JSON with blood group """ if 'image' not in request.files: return jsonify({'error': 'No image file provided'}), 400 file = request.files['image'] if file.filename == '': return jsonify({'error': 'No file selected'}), 400 if not allowed_file(file.filename): return jsonify({'error': 'File type not allowed. Use PNG, JPG, or JPEG'}), 400 # Save file filename = secure_filename(file.filename) unique_name = str(uuid.uuid4()) + '_' + filename file_path = os.path.join(app.config['UPLOAD_FOLDER'], unique_name) file.save(file_path) try: result = detect_blood_group(file_path) return jsonify({'success': True, **result}) except Exception as e: return jsonify({'error': f'Processing failed: {str(e)}'}), 500

# ------------------- Web Interface ------------------- @app.route('/', methods=['GET', 'POST']) def index(): if request.method == 'POST': if 'image' not in request.files: return render_template('index.html', error='No file uploaded') file = request.files['image'] if file.filename == '': return render_template('index.html', error='No file selected') if not allowed_file(file.filename): return render_template('index.html', error='File type not allowed. Use PNG, JPG, or JPEG') # Save file filename = secure_filename(file.filename) unique_name = str(uuid.uuid4()) + '_' + filename file_path = os.path.join(app.config['UPLOAD_FOLDER'], unique_name) file.save(file_path) try: result = detect_blood_group(file_path) return render_template('index.html', result=result) except Exception as e: return render_template('index.html', error=f'Processing failed: {str(e)}') return render_template('index.html')

@app.route('/health', methods=['GET']) def health(): return jsonify({'status': 'ok'})

if __name__ == '__main__': app.run(debug=True, host='0.0.0.0', port=5000) ```

---

### 4. `templates/index.html` (Simple Web Interface)

```html Blood Group Detection * { margin: 0; padding: 0; box-sizing: border-box; } body { font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif; background: linear-gradient(135deg, #667eea 0%, #764ba2 100%); min-height: 100vh; display: flex; justify-content: center; align-items: center; padding: 20px; } .container { background: white; padding: 40px; border-radius: 20px; box-shadow: 0 15px 35px rgba(0,0,0,0.2); max-width: 500px; width: 100%; } h1 { color: #333; text-align: center; margin-bottom: 10px; font-size: 28px; } .subtitle { color: #777; text-align: center; margin-bottom: 30px; font-size: 14px; } .upload-form { display: flex; flex-direction: column; gap: 15px; } .file-input { padding: 15px; border: 2px dashed #ddd; border-radius: 10px; cursor: pointer; text-align: center; transition: all 0.3s; background: #f8f9fa; } .file-input:hover { border-color: #667eea; background: #f0f2ff; } .file-input input { display: none; } .file-label { color: #555; } .btn { background: #667eea; color: white; border: none; padding: 15px; border-radius: 10px; font-size: 16px; font-weight: 600; cursor: pointer; transition: all 0.3s; } .btn:hover { background: #5a6fd6; transform: translateY(-2px); } .result-box { margin-top: 30px; padding: 20px; border-radius: 10px; background: #f0f4ff; border-left: 4px solid #667eea; } .blood-group { font-size: 36px; font-weight: 700; color: #667eea; text-align: center; margin: 10px 0; } .agg-details { display: flex; justify-content: space-around; margin-top: 15px; } .agg-item { text-align: center; padding: 10px; background: white; border-radius: 8px; box-shadow: 0 2px 8px rgba(0,0,0,0.1); } .agg-item .label { font-size: 12px; color: #888; margin-bottom: 5px; } .agg-item .value { font-weight: 600; color: #333; } .error { margin-top: 20px; padding: 15px; background: #fee; color: #c00; border-radius: 8px; border: 1px solid #fcc; text-align: center; } .api-info { margin-top: 30px; padding-top: 20px; border-top: 1px solid #eee; text-align: center; color: #999; font-size: 13px; } 🩸 Blood Group Detector Upload an image of a blood typing card to detect the blood group

📁 Click to select or drag image 🔬 Detect Blood Group {% if result %} Detected Blood Group {{ result.blood_group }} Anti-A {{ result.agglutination.anti_A }} Anti-B {{ result.agglutination.anti_B }} Anti-D (Rh) {{ result.agglutination.anti_D }} {% endif %} {% if error %} ⚠️ {{ error }} {% endif %} API Endpoint: POST /api/detect (multipart/form-data with field "image") Returns JSON: {"blood_group": "A+", "agglutination": {...}} ```

---

### 5. How to Run

1. Install dependencies: ```bash pip install -r requirements.txt ```

2. Run the application: ```bash python app.py ```

3. Access the web interface at: `http://localhost:5000` 4. API testing (using curl): ```bash curl -X POST -F "image=@path/to/blood_test.jpg" http://localhost:5000/api/detect ```

---

### 6. Important Notes

- The code assumes a standard blood typing card with 3 wells: Anti-A, Anti-B, Anti-D (from left to right). Upload images should be clear, well-lit, and oriented correctly. - Agglutination detection is heuristic - it works best with clean images showing clear clumping. For production, you would use a trained CNN or object detection model on labeled images. - The threshold for agglutination (count > 3 or density > 0.05) can be adjusted based on your specific image characteristics. - Always test with your own sample images to calibrate thresholds.

This provides a functional end-to-end solution with both a user-friendly web interface and a REST API.

Détection du Groupe Sanguin à partir d'une Image - API/Website Python

Contenu du PromptEnregistrer

Connectez-vous pour voir le prompt complet

Utilisation

Références

Discussion