数据魔术师小白零基础实现简单人脸识别

import cv2  # opencv
import os
def generate(dirname):
    face_cascade = cv2.CascadeClassifier(r'C:Users86155AppDataLocalProgramsPythonPython37Libsite-packagescv2datahaarcascade_frontalface_default.xml'
    )#导入OpenCV已经训练好的数据文件

    if (not os.path.isdir(dirname)):
        os.makedirs(dirname)
    camera = cv2.VideoCapture(0)  #参数为0表示打开笔记本摄像头，参数为路径表示打开视频
    count = 0   #计数变量
    while (True):
        ret, frame = camera.read()  #camera.read()按帧读取视频，ret,frame是获cap.read()方法的两个返回值。其中ret是布尔值，如果读取帧是正确的则返回True，如果文件读取到结尾，它的返回值就为False。frame就是每一帧的图像，是个三维矩阵
        gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)  #色彩空间的转化 生成灰度图
        faces = face_cascade.detectMultiScale(gray,1.3,5)  #scaleFactor表示每次图像尺寸减小的比例  minNeighbors表示每一个目标至少要被检测到5次才算是真的目标
        for (x, y, w, h) in faces:
            img = cv2.rectangle(frame, (x, y), (x + w, y + h), (255, 0, 0),2) #在frame上以两点连成对角线确定蓝色矩形框
            f = cv2.resize(gray[y:y + h, x:x + w],(200, 200))  #将人脸部分分辨率调整为200*200
            cv2.imwrite(dirname + '/%s.pgm' % str(count), f)  #将图片写入硬盘
            print(count)
            count += 1

        cv2.imshow("camera", frame)  #在摄像头上显示获取的人脸部分图片
        if cv2.waitKey(100) & 0xff == ord("q"):  #按q退出
            break
        elif count > 20:  #拍摄二十张后退出
            break

    camera.release()
    cv2.destroyAllWindows()  #关闭相机，释放资源

if __name__ == "__main__":
    generate(r"C:Users86155Desktop123editor")  #传入图片保存路径，路径中不要出现中文

import sys
import os.path
if __name__ == "__main__":
    BASE_PATH = r"C:Users86155Desktop123"

    SEPARATOR = ";"

    fh = open(r"C:Users86155Desktop123at.text", 'w')

    label = 0

    for dirname, dirnames, filenames in os.walk(BASE_PATH):
        for subdirname in dirnames:
            subject_path = os.path.join(dirname, subdirname)
            for filename in os.listdir(subject_path):
                abs_path = "%s/%s" % (subject_path, filename)
                print("%s%s%d" % (abs_path, SEPARATOR, label))
                fh.write(abs_path)
                fh.write(SEPARATOR)
                fh.write(str(label))
                fh.write("n")
            label = label + 1
    fh.close()

import os
import sys
import numpy as np
import cv2


def read_images(path, sz=None):
    c = 0  #标签号
    X, y = [], []
    for dirname, dirnames, filenames in os.walk(path):
        for subdirname in dirnames:
            subject_path = os.path.join(dirname, subdirname)
            for filename in os.listdir(subject_path):
                try:
                    if not filename.endswith('.pgm'):
                        continue
                    filepath = os.path.join(subject_path, filename)
                    im = cv2.imread(
                        filepath, cv2.IMREAD_GRAYSCALE
                    )  #读取图片后已多维数组的形式保存图片信息，前两维表示图片的像素坐标，最后一维表示图片的通道索引
                    if sz is not None:
                        im = cv2.resize(im, (200, 200))
                    X.append(np.asarray(im, dtype=np.uint8))
                    y.append(c)
                except:
                    print("Unexpected error:", sys.exc_info()[0])
            c = c + 1
    return [X, y]


def face_rec(img_path):
    names = ['editor', 'roommate']     #标签对应的人名
    [X, y] = read_images(img_path)
    y = np.asarray(y, dtype=np.int32)
    model = cv2.face.EigenFaceRecognizer_create()  #特征脸人脸识别算法
    model.train(np.asarray(X), np.asarray(y))  #传入特征（原始图像）和标签进行训练
    camera = cv2.VideoCapture(0)  #打开摄像头
    face_cascade = cv2.CascadeClassifier(
        r'C:Users86155AppDataLocalProgramsPythonPython37Libsite-packagescv2datahaarcascade_frontalface_default.xml'
    )  #读入级联分类器
    while (True):
        read, img = camera.read()  #第一个返回值为boolen值
        faces = face_cascade.detectMultiScale(img, 1.3, 5)
        for (x, y, w, h) in faces:
            img = cv2.rectangle(img, (x, y), (x + w, y + h), (255, 0, 0), 2)
            gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
            roi = gray[x:x + w, y:y + h]   #重复第一步图像获取
            try:
                roi = cv2.resize(
                    roi, (200, 200), interpolation=cv2.INTER_LINEAR)  #处理刚刚回去的要识别的人脸
                params = model.predict(roi)  #进行识别  返回最接近的标签和置信度（数值越小，准确性越高）
                print("Label: %s, Confidence: %.2f" % (params[0], params[1]))
                cv2.putText(img, names[params[0] - 1], (x, y - 20),
                            cv2.FONT_HERSHEY_SIMPLEX, 1, 255, 2)   #将识别出的人名信息显示在图像上
            except:
                continue
        cv2.imshow("camera", img)
        if cv2.waitKey(1000 // 12) & 0xff == ord('q'):
            break  #按q结束
    cv2.destroyAllWindows()


if __name__ == "__main__":
    face_rec(r'C:Users86155Desktop123')   #传入原始图像的父文件夹