face recog 2.py

import cv2
import numpy as np
from os import listdir
from os.path import isdir, isfile, join

face_classifier = cv2.CascadeClassifier('haarcascade_frontalface_default.xml')    

# 사용자 얼굴 학습
def train(name):
    data_path = 'faces/' + name + '/'
    face_pics = [f for f in listdir(data_path) if isfile(join(data_path,f))]
    Training_Data, Labels = [], []
    
    for i, files in enumerate(face_pics):
        image_path = data_path + face_pics[i]
        images = cv2.imread(image_path, cv2.IMREAD_GRAYSCALE)
        if images is None:
            continue    
        Training_Data.append(np.asarray(images, dtype=np.uint8))
        Labels.append(i)
    if len(Labels) == 0:
        print("인식 불가.")
        return None
    Labels = np.asarray(Labels, dtype=np.int32)
     # 모델 생성
    model = cv2.face.LBPHFaceRecognizer_create()
    model.train(np.asarray(Training_Data), np.asarray(Labels))
    print(name + " : 모델 학습 완료 !!")

    return model

#여러 사용자 학습
def trains():
    data_path = 'faces/'
    model_dirs = [f for f in listdir(data_path) if isdir(join(data_path,f))]
    models = {}
    for model in model_dirs:
        print('model :' + model)
        result = train(model)
        if result is None:
            continue

        print('model2 :' + model)
        models[model] = result
    return models           # 학습된 모델들 리턴

#얼굴 검출
def face_detector(img, size = 0.5):
        gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
        faces = face_classifier.detectMultiScale(gray,1.3,5)
        if faces is():
            return img,[]
        for(x,y,w,h) in faces:
            cv2.rectangle(img, (x,y),(x+w,y+h),(0,255,255),2)
            roi = img[y:y+h, x:x+w]
            roi = cv2.resize(roi, (200,200))
        return img,roi   



def run(models):    
    frame = cv2.imread('./faces/ppp.jpg')
    image, face = face_detector(frame)
  
    try:            
        min_score = 999       
        min_score_name = ""  
        face = cv2.cvtColor(face, cv2.COLOR_BGR2GRAY)
    
        for key, model in models.items():
            result = model.predict(face)
            print(result)                
            if min_score > result[1]:
                min_score = result[1]
                min_score_name = key
        
        if min_score < 500: 
            confidence = int(100*(1-(min_score)/300))
            #display_string = '  Corretion rate '+ str(confidence) + '%'
        #cv2.putText(image,display_string,(100,120), cv2.FONT_HERSHEY_COMPLEX,1,(250,120,255),2)
        print("cc",confidence)
        if confidence > 85:
            print("Unlocked : " + min_score_name)
            #cv2.imshow('Face Cropper', image)
            print("Unlocked")
        else:
            print("Locked"+"  : not admin user")
            #cv2.imshow('Face Cropper', image)
            print("locked")
    except:
        #cv2.putText(image, "Face Not Found", (500, 500), cv2.FONT_HERSHEY_COMPLEX, 1, (255, 0, 0), 2)
        #cv2.imshow('Face Cropper', image)
        print("face not found")
        pass
    
    cv2.destroyAllWindows()

if __name__ == "__main__":
    models = trains()
    run(models)