基于OpenCV的人脸检测，单人脸检测和合照中多人脸检测，Python实现

ゝ一世哀愁。 2023-05-29 06:50 16阅读 0赞

# 基于OpenCV的人脸检测，单人脸检测和合照中多人脸检测，Python实现 #

本例基于opencv4。Python 3.7.4。

Python安装OpenCV4.0的命令：

pip install opencv-python

安装完成后：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA_size_16_color_FFFFFF_t_70][]

需要到OpenCV官方GitHub下载人脸检测的分类器：[https://github.com/opencv/opencv/tree/master/data/haarcascades][https_github.com_opencv_opencv_tree_master_data_haarcascades]

然后把分类器和所要检测的图片放到一定目录文件夹下加载。我把分类器和图片放在项目根目录下：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA_size_16_color_FFFFFF_t_70 1][]

Python代码：

import cv2 as cv
    
    imagepath = "./pic1.jpg"
    
    image = cv.imread(imagepath)
    
    gray = cv.cvtColor(image, cv.COLOR_BGR2GRAY)
    
    face_cascade = cv.CascadeClassifier('./haarcascade_frontalface_default.xml')
    
    faces = face_cascade.detectMultiScale(
        gray,
        scaleFactor=1.2,
        minNeighbors=3,
        minSize=(70, 70),
        maxSize=(160, 160),
        flags=cv.IMREAD_GRAYSCALE
    )
    
    print("检测到{0}个人脸".format(len(faces)))
    
    for (x, y, w, h) in faces:
        cv.rectangle(image, (x, y), (x + w, y + w), (0, 0, 255), 2)
    
    cv.imshow("image", image)
    
    cv.waitKey(0)

检测计算机视觉经典照片lena，原图：

![20191105130647189.jpg][]

运行代码后，红框标记出检测到的人脸：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA_size_16_color_FFFFFF_t_70 2][]

程序检测到1张人脸。

换一张合影照片，看看能否检测到群体像中的人脸，原图：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA_size_16_color_FFFFFF_t_70 3][]

运行代码后，红框标记出检测到的人脸：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA_size_16_color_FFFFFF_t_70 4][]

程序检测到五张人脸。

再来一张，原图：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA_size_16_color_FFFFFF_t_70 5][]

运行程序：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA_size_16_color_FFFFFF_t_70 6][]

程序检测到2个人脸，并用红框标记出人脸范围。

[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA_size_16_color_FFFFFF_t_70]: https://img-blog.csdnimg.cn/20191105124252932.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA==,size_16,color_FFFFFF,t_70
[https_github.com_opencv_opencv_tree_master_data_haarcascades]: https://github.com/opencv/opencv/tree/master/data/haarcascades
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA_size_16_color_FFFFFF_t_70 1]: https://img-blog.csdnimg.cn/20191105124411436.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA==,size_16,color_FFFFFF,t_70
[20191105130647189.jpg]: https://img-blog.csdnimg.cn/20191105130647189.jpg
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA_size_16_color_FFFFFF_t_70 2]: https://img-blog.csdnimg.cn/20191105130719108.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA==,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA_size_16_color_FFFFFF_t_70 3]: https://img-blog.csdnimg.cn/20191105130759442.jpg?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA==,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA_size_16_color_FFFFFF_t_70 4]: https://img-blog.csdnimg.cn/20191105130902591.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA==,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA_size_16_color_FFFFFF_t_70 5]: https://img-blog.csdnimg.cn/2019110513121410.jpg?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA==,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA_size_16_color_FFFFFF_t_70 6]: https://img-blog.csdnimg.cn/20191105131326222.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA==,size_16,color_FFFFFF,t_70