🎭 التعرف على مشاعر الوجه
ارفع صورة للتعرف على المشاعر الموجودة فيها
😠 Angry
😤 Contempt
🤢 Disgust
😨 Fear
😊 Happy
😐 Neutral
😢 Sad
😲 Surprise
API Endpoints:
- POST /predict - رفع صورة للتنبؤ
- GET /emotions - قائمة المشاعر المدعومة
- GET /health - فحص حالة الخدمة
"""
return HTMLResponse(content=html_content)
@app.get("/health")
async def health_check():
return {
"status": "healthy",
"model_loaded": model is not None,
"supported_emotions": len(EMOTIONS)
}
@app.get("/emotions")
async def get_emotions():
return {
"emotions": EMOTIONS,
"count": len(EMOTIONS),
"description": "8 مشاعر أساسية يمكن التعرف عليها"
}
@app.post("/predict")
async def predict_emotion(file: UploadFile = File(...)):
if model is None:
raise HTTPException(status_code=503, detail="Model not loaded")
try:
contents = await file.read()
image = Image.open(io.BytesIO(contents))
image_np = np.array(image)
if len(image_np.shape) == 3 and image_np.shape[2] == 3:
image_np = cv2.cvtColor(image_np, cv2.COLOR_RGB2BGR)
results = model.predict(image_np, verbose=False, conf=0.25)
predictions = []
for result in results:
if result.boxes is not None and len(result.boxes) > 0:
for box in result.boxes:
conf = float(box.conf[0])
cls = int(box.cls[0])
bbox = box.xyxy[0].tolist()
emotion_name = EMOTIONS.get(cls, f"unknown_{cls}")
# احتمالات باقي المشاعر (لو متوفرة)
probs_dict = {}
if hasattr(box, 'probs') and box.probs is not None:
probs = box.probs[0].tolist()
for i, p in enumerate(probs):
probs_dict[EMOTIONS.get(i, f"unknown_{i}")] = round(p, 4)
else:
probs_dict = {emotion_name: round(conf, 4)}
predictions.append({
"emotion": emotion_name,
"confidence": round(conf, 4),
"bbox": {
"x1": int(bbox[0]),
"y1": int(bbox[1]),
"x2": int(bbox[2]),
"y2": int(bbox[3])
},
"class_id": cls,
"probabilities": probs_dict
})
predictions.sort(key=lambda x: x["confidence"], reverse=True)
return JSONResponse(content={
"success": True,
"image_name": file.filename,
"predictions": predictions,
"faces_detected": len(predictions),
"model_info": {
"model_file": "best.pt",
"supported_emotions": list(EMOTIONS.values())
}
}, status_code=200)
except Exception as e:
return JSONResponse(content={
"success": False,
"error": str(e),
"message": "فشل في تحليل الصورة"
}, status_code=500)
@app.post("/predict-base64")
async def predict_emotion_base64(image_data: str = Form(...)):
if model is None:
raise HTTPException(status_code=503, detail="Model not loaded")
try:
image_bytes = base64.b64decode(image_data)
image = Image.open(io.BytesIO(image_bytes))
image_np = np.array(image)
if len(image_np.shape) == 3 and image_np.shape[2] == 3:
image_np = cv2.cvtColor(image_np, cv2.COLOR_RGB2BGR)
results = model.predict(image_np, verbose=False, conf=0.25)
predictions = []
for result in results:
if result.boxes is not None and len(result.boxes) > 0:
for box in result.boxes:
conf = float(box.conf[0])
cls = int(box.cls[0])
bbox = box.xyxy[0].tolist()
probs_dict = {}
if hasattr(box, 'probs') and box.probs is not None:
probs = box.probs[0].tolist()
for i, p in enumerate(probs):
probs_dict[EMOTIONS.get(i, f"unknown_{i}")] = round(p, 4)
else:
emotion_name = EMOTIONS.get(cls, f"unknown_{cls}")
probs_dict = {emotion_name: round(conf, 4)}
predictions.append({
"emotion": EMOTIONS.get(cls, f"unknown_{cls}"),
"confidence": round(conf, 4),
"bbox": {
"x1": int(bbox[0]),
"y1": int(bbox[1]),
"x2": int(bbox[2]),
"y2": int(bbox[3])
},
"probabilities": probs_dict
})
predictions.sort(key=lambda x: x["confidence"], reverse=True)
return {
"success": True,
"predictions": predictions,
"faces_detected": len(predictions)
}
except Exception as e:
raise HTTPException(status_code=500, detail=f"Base64 prediction error: {str(e)}")
if __name__ == "__main__":
uvicorn.run(app, host="0.0.0.0", port=7860)