OpenCV Python人脸识别仅一个人脸识别

Question

嗨，大家好，当我运行下面的代码时，我们正在开发一个人脸识别应用程序，正如您在照片中看到的那样，它仅在一个人脸（红色正方形）中起作用，它不扫描转换数据中的其他人脸，我猜我的预测函数只运行一次。 不要在循环中。

处理后的图片： LINK

 # coding: utf-8
 import cv2
 import os
 import numpy as np
 suclular = ["Bilinmeyen", "Veli Eroglu", "Ali Eroglu"]


 def detect_face(img):
     # ALGORİMA için Gri Yapıyoruz.
     gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
     # yüz tanımlama için geereken haarcascade
     face_cascade = cv2.CascadeClassifier(
         'opencv-files/lbpcascade_frontalface.xml')
     faces = face_cascade.detectMultiScale(
         gray, scaleFactor=1.2, minNeighbors=5)  # YÜZ TANIMLAMA
     for (x, y, w, h) in faces:
         img = cv2.rectangle(img, (x, y), (x + w, y + h), (255, 0, 0), 2)
     if (len(faces) == 0):
         return None, None  # Yuz bulunamazsa...
     (x, y, w, h) = faces[0]
     return gray[y:y + w, x:x + h], faces[0]

 face_recognizer = cv2.face.LBPHFaceRecognizer_create()
 face_recognizer.train(faces, np.array(labels))


 def draw_rectangle(img, rect):
     (x, y, w, h) = rect
     cv2.rectangle(img, (x, y), (x + w, y + h), (0, 0, 255), 2)


 def draw_text(img, text, x, y):
     cv2.putText(img, text, (x, y), cv2.FONT_HERSHEY_PLAIN, 1.5, (0, 0, 255), 2)


 def predict(test_img):
     img = test_img.copy()
     face, rect = detect_face(img)
     label, confidence = face_recognizer.predict(face)
     print(confidence)
     label_text = suclular[label]
     if confidence > 42 and confidence < 70:
         label_text = "Tespit Edilemedi."
         print(label_text)
     elif confidence > 70:
         label_text = "Bilinmiyor"
     draw_rectangle(img, rect)
     draw_text(img, label_text, rect[0], rect[1] - 5)

     return img


 print("Predicting images...")
 test_img1 = cv2.imread("test-data/test8jpg.jpg")
 predicted_img1 = predict(test_img1)
 print("Prediction complete")
 cv2.imshow("SONUC", cv2.resize(predicted_img1, (400, 500)))
 cv2.waitKey(0)
 cv2.destroyAllWindows()
 cv2.waitKey(1)
 cv2.destroyAllWindows()

Answer 1

您的预测应该在for循环中使用...即使您遍历每张脸并为每张脸制作一个矩形，您也只能从detect_face函数返回一张脸，这是最后一张脸。这个：

def predict_face(img):
     # ALGORİMA için Gri Yapıyoruz.
     gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
     # yüz tanımlama için geereken haarcascade
     face_cascade = cv2.CascadeClassifier(
         'opencv-files/lbpcascade_frontalface.xml')
     faces = face_cascade.detectMultiScale(
         gray, scaleFactor=1.2, minNeighbors=5)  # YÜZ TANIMLAMA
     detected_faces = []
     i = 0
     for (x, y, w, h) in faces:
         img = cv2.rectangle(img, (x, y), (x + w, y + h), (255, 0, 0), 2)
         detected_face = gray[y:y + w, x:x + h]
         label, confidence = face_recognizer.predict(detected_face)  # Prediction inside for loop
         draw_rectangle(img, faces[i])    # draw the red rectangles for every predicted face
         draw_text(img, label, x, y - 5)  # draw the predicted label on top of the box
         i += 1

• 参见for循环中的第三行...我的预测在for循环内
• 甚至在for循环中绘制矩形
• 无需存储矩形并再次遍历它们即可绘制它们
• 在for循环中绘制预测
• 因此，您可以在一个功能中完成所有操作

Answer 2

为什么要将原始图像传递给predict()函数？ 一旦检测到两个脸部，就应该将它们中的每一个（ 使用OpenCV函数从原始图像中提取 ）传递给预测函数。 这样，您的算法将可以工作并且更快。

祝好运！