本文的重点内容为opencv+flask+face_recognition
先上开源代码,https://github.com/Kr1s77/flask-video-streaming-recorder 作者通过opencv+flask完成了读取摄像头,并且在HTML页面显示的功能,很多场景都会用到,本文是分享二次开发人脸识别。
源码的main.py中源码app.run(threaded=True, host="0.0.0.0"),默认了port=5000,端口是可以改变的,如下app.run(threaded=True, host="0.0.0.0",port=5002)
改完之后运行python main.py,打开浏览器访问http://localhost:5002/login 就可以看到登录页面啦!(作者忘了提示这步)
Username: admin
Password: admin
本文用开源的face_recognition(如果没有安装就pip install face_recognition)
只需要修改一个文件:controller/utils/camera.py,下面直接给完整代码
import cv2
import threading
import face_recognition
class RecordingThread(threading.Thread):
def __init__(self, name, camera):
threading.Thread.__init__(self)
self.name = name
self.isRunning = True
self.cap = camera
fourcc = cv2.VideoWriter_fourcc(*'MJPG')
self.out = cv2.VideoWriter('./static/video.avi', fourcc, 20.0, (640, 480))
def run(self):
while self.isRunning:
ret, frame = self.cap.read()
if ret:
self.out.write(frame)
self.out.release()
def stop(self):
self.isRunning = False
def __del__(self):
self.out.release()
class VideoCamera(object):
def __init__(self):
# 打开摄像头, 0代表笔记本内置摄像头
self.cap = cv2.VideoCapture(0)
#初始化人脸
obama_img = face_recognition.load_image_file("obama.jpg")
self.obama_face_encoding = face_recognition.face_encodings(obama_img)[0]
self.face_locations = []
self.face_encodings = []
self.face_names = []
self.process_this_frame = True
# 初始化视频录制环境
self.is_record = False
self.out = None
# 视频录制线程
self.recordingThread = None
# 退出程序释放摄像头
def __del__(self):
self.cap.release()
def get_frame(self):
ret, frame = self.cap.read()
if ret:
# 将视频帧调整为1/4大小,以加快脸部识别处理
small_frame = cv2.resize(frame,(0,0),fx=0.25, fy=0.25)
# 将图像从BGR颜色(OpenCV使用的)转换为RGB颜色(face_recognition使用)
rgb_small_frame = small_frame[:, :, ::-1]
if self.process_this_frame:
# 查找当前视频帧中的所有面部和脸部编码
self.face_locations = face_recognition.face_locations(small_frame)
self.face_encodings = face_recognition.face_encodings(small_frame, self.face_locations)
self.face_names = []
#处理多张人脸的情况
for self.face_encoding in self.face_encodings:
# 查看脸部是否与已知脸部相匹配(S)
match = face_recognition.compare_faces([self.obama_face_encoding], self.face_encoding)
if match[0]:
name = "obama"
else:
name = "unkonwn"
self.face_names.append(name)
self.process_this_frame = not self.process_this_frame
#对有人脸的图进行处理
for (top, right, bottom, left), name in zip(self.face_locations, self.face_names):
# 自从我们检测到的框架缩放到1/4尺寸后,缩放后面的位置
top *= 4
right *= 4
bottom *= 4
left *= 4
# 在脸上画一个方框
cv2.rectangle(frame, (left, top), (right, bottom), (0, 0, 255), 2)
# 在脸部下面画一个名字
cv2.rectangle(frame, (left, bottom - 35), (right, bottom), (0, 0, 255), 2)
font = cv2.FONT_HERSHEY_DUPLEX
cv2.putText(frame, name, (left+6, bottom-6), font, 1.0, (255, 255, 255), 1)
# 因为opencv读取的图片并非jpeg格式,因此要用motion JPEG模式需要先将图片转码成jpg格式图片
ret, jpeg = cv2.imencode('.jpg', frame)
# 视频录制
if self.is_record:
if self.out == None:
fourcc = cv2.VideoWriter_fourcc(*'MJPG')
self.out = cv2.VideoWriter('./static/video.avi', fourcc, 20.0, (640, 480))
ret, frame = self.cap.read()
if ret:
self.out.write(frame)
else:
if self.out != None:
self.out.release()
self.out = None
return jpeg.tobytes()
else:
return None
def start_record(self):
self.is_record = True
self.recordingThread = RecordingThread("Video Recording Thread", self.cap)
self.recordingThread.start()
def stop_record(self):
self.is_record = False
if self.recordingThread != None:
self.recordingThread.stop()
obama_img = face_recognition.load_image_file("obama.jpg") ,我这里只加了一张奥巴马的图片作为人脸库,这张图片你可以拍一张自己的照片替换,放在目录下就好了,没有照片会报错的。
加人脸识别的地方,可以改成调用某些平台的AI图片处理API,自行封装哈。