简单创建一个认识自己的代码1：数据收集

2019年10月15号

作者：杨轮飞

将会使用到python中的opencv、dlib、sklearn库。

1.先通过opencv+dlib调用摄像头对人脸进行截图保存为数据集

• 先导入库

import cv2
import dlib
import numpy

• 写一个使用dlib检测人脸后对图像的函数，用于输出框出人脸的图像与人脸区域坐上角点和右下角

def dlib_find_face(img, dets):
    font = cv2.FONT_HERSHEY_SCRIPT_SIMPLEX
    minx, miny, maxx, maxy = 640, 640, 0, 0
    for k, d in enumerate(dets):
        shape = predictor(img, d)
        #(x, y, w, h) = face_utils.rect_to_bb(rect)
        landmark = numpy.matrix([[p.x, p.y] for p in shape.parts()])
        for idx, point in enumerate(landmark):
            x, y = (point[0, 0], point[0, 1])
            cv2.putText(img, "*", (x, y), font,0.3,(0, 255, 0))
            if x > maxx: maxx = x;
            if x < minx: minx = x;
            if y > maxy: maxy = y;
            if y < miny: miny = y;
    cv2.rectangle(img, (minx - 10, miny - 10), (maxx + 10, maxy + 10), (0, 255, 0), 2)
    return img, (minx, miny), (maxx, maxy)

• 打开摄像头，对图像进行dlib人脸检测，然后将人脸区域保存下来。当保存了32张人脸时推出程序

if __name__=="__main__":
    predictor_path = "F:\qq\shape_predictor_68_face_landmarks.dat"
    detector = dlib.get_frontal_face_detector()
    predictor = dlib.shape_predictor(predictor_path)

    cap = cv2.VideoCapture(0)
    number = 1
    name = "wzy"
    while True:
        sucess, img = cap.read()
        temp = img.copy()
        temp_gray = cv2.cvtColor(temp, cv2.COLOR_RGB2GRAY)
        dets = detector(temp_gray, 1)
        k = cv2.waitKey(1)
        if k == 27:
            break
        if len(dets) == 1:
            temp, min, max = dlib_find_face(temp, dets)
            if k == 32:
                face = img[min[1]:max[1], min[0]:max[0]]
                face = cv2.resize(face, (64, 64))
                cv2.imwrite("face_image\\"+name+"\\"+name+str(number)+".jpg", face)
                print("保存成功："+name+str(number)+".jpg")
                number += 1
                if number == 33: break;
        cv2.imshow("face", temp)
    cap.release()
    cv2.destroyAllWindows()

得到数据集后第一步算是结束啦~

接下来就是对我们的数据集进行训练了~