mood.py 文件源码

def main_func():
    img_path='snap.jpg' # THE PATH OF THE IMAGE TO BE ANALYZED

    font=cv2.FONT_HERSHEY_DUPLEX
    emotions = ["anger", "happy", "sadness"] #Emotion list
    clahe=cv2.createCLAHE(clipLimit=2.0,tileGridSize=(8,8)) # Histogram equalization object
    face_det=dlib.get_frontal_face_detector()
    land_pred=dlib.shape_predictor("data/DlibPredictor/shape_predictor_68_face_landmarks.dat")



    SUPPORT_VECTOR_MACHINE_clf2 = joblib.load('data/Trained_ML_Models/SVM_emo_model_7.pkl')
    # Loading the SVM model trained earlier in the path mentioned above.



    pred_data=[]
    pred_labels=[]

    a=crop_face(img_path)
    img=cv2.imread(a)
    gray=cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)
    clahe_gray=clahe.apply(gray)
    landmarks_vec = get_landmarks(clahe_gray,face_det,land_pred)

    #print(len(landmarks_vec))
    #print(landmarks_vec)

    if landmarks_vec == "error":
        pass
    else:
        pred_data.append(landmarks_vec)
    np_test_data = np.array(pred_data)
    a=SUPPORT_VECTOR_MACHINE_clf2.predict(pred_data)
    #cv2.putText(img,'DETECTED FACIAL EXPRESSION : ',(8,30),font,0.7,(0,0,255),2,cv2.LINE_AA)
    #l=len('Facial Expression Detected : ')
    #cv2.putText(img,emotions[a[0]].upper(),(150,60),font,1,(255,0,0),2,cv2.LINE_AA)
    #cv2.imshow('test_image',img)
    #print(emotions[a[0]])


    cv2.waitKey(0)
    cv2.destroyAllWindows()
    return emotions[a[0]]