1. github地址
https://github.com/ipazc/mtcnn
2. 安装
Currently it is only supported Python3.4 onwards. It can be installed through pip:
$ pip install mtcnn
This implementation requires OpenCV>=4.1 and Keras>=2.0.0 (any Tensorflow supported by Keras will be supported by this MTCNN package). If this is the first time you use tensorflow, you will probably need to install it in your system:
$ pip install tensorflow==1.9.0
$ pip install keras==2.2.0
3. 环境配置
我的版本:
python 3.6.5
tensorflow 1.9.0
Keras 2.2.0
其他可参考 https://www.cnblogs.com/carle-09/p/11661261.html
但是keras 版本必须大于 2.2
4. 实例代码 人脸识别
4.1 库
from mtcnn import MTCNN
detector = MTCNN()
import numpy as np
import cv2
4.2 检测函数
- 输入是一个 height * weight * 3的图片 (三维数组), 可以用cv2.imread() 读取
- 输出是 一个数组, 每个元素是 height * weight * 3 , 表示的是 面部的 三维数组, 可以用cv2.imwrite()保存
def detect_face (img):
# img = height * weight * 3 (RGB)
shapes = np.array(img).shape
# face_arr
face_arr = []
# 得到高度和宽度 做异常处理
height = shapes[0]
weight = shapes[1]
#检测人脸
detect_result = detector.detect_faces(img)
#如果检测不到人脸、 返回空
if len(detect_result )== 0:
return []
else :
for item in detect_result:
box = item['box']
#因为预测是 给出左上角的点座标【0,1】 以及 长宽【2,3】 所以需要转换
top = box[1]
buttom = box[1] + box[3]
left = box[0]
right = box[0] + box[2]
#因为左上角的点可能会在图片范围外 所以要异常处理
if top < 0:
top = 0
if left < 0:
left = 0
if buttom > height:
buttom = height
if right > weight:
right = weight
face_arr.append(img[top: buttom, left: right])
return face_arr
4.3 实例
import cv2
img = cv2.imread('test1.png')
# jpg保存
face_arr = detect_face(img)
if not len(face_arr) == 0:
count = 0
# 把脸部图片全部进行 写出
for item in face_arr:
count = count + 1
cv2.imwrite( "face_" + str(count) + '.png', item ) # 写入图片 jpg有损压缩,png无损压缩
else:
print("No face is detected")