OpenCV从入门到精通实战(九)——基于dlib的疲劳监测 ear计算

本文主要是介绍OpenCV从入门到精通实战(九)——基于dlib的疲劳监测 ear计算，希望对大家解决编程问题提供一定的参考价值，需要的开发者们随着小编来一起学习吧！

本文实现Python库d和OpenCV来实现眼部闭合检测，主要用于评估用户是否眨眼。

步骤一：导入必要的库和设置参数

首先，代码导入了必要的Python库，如dlib、OpenCV和scipy。通过argparse设置了输入视频和面部标记预测器的参数。

from scipy.spatial import distance as dist
from collections import OrderedDict
import numpy as np
import argparse
import time
import dlib
import cv2

步骤二：定义面部关键点索引

使用OrderedDict定义了包含68个点的面部关键点，用于眼部分析。

FACIAL_LANDMARKS_68_IDXS = OrderedDict([("mouth", (48, 68)),("right_eyebrow", (17, 22)),("left_eyebrow", (22, 27)),("right_eye", (36, 42)),("left_eye", (42, 48)),("nose", (27, 36)),("jaw", (0, 17))
])

步骤三：定义眼部闭合比率（EAR）函数

此函数计算眼部的纵横比，用于判断眼睛是否闭合。

def eye_aspect_ratio(eye):A = dist.euclidean(eye[1], eye[5])B = dist.euclidean(eye[2], eye[4])C = dist.euclidean(eye[0], eye[3])ear = (A + B) / (2.0 * C)return ear

步骤四：读取视频并初始化检测器

加载面部检测器和预测器，读取视频流。

detector = dlib.get_frontal_face_detector()
predictor = dlib.shape_predictor(args["shape_predictor"])
vs = cv2.VideoCapture(args["video"])

步骤五：遍历视频帧，检测和分析

对视频的每一帧进行处理，检测人脸，提取眼部关键点，并计算EAR。

while True:frame = vs.read()[1]gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)rects = detector(gray, 0)

步骤六：闭眼检测逻辑

分析EAR值，累计闭眼帧数，并计算总的眨眼次数。

if ear < EYE_AR_THRESH:COUNTER += 1
else:if COUNTER >= EYE_AR_CONSEC_FRAMES:TOTAL += 1COUNTER = 0

步骤七：显示结果

在视频帧上显示眨眼次数和当前EAR值，同时绘制眼部区域。

cv2.putText(frame, "Blinks: {}".format(TOTAL), (10, 30), cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)
cv2.drawContours(frame, [leftEyeHull], -1, (0, 255, 0), 1)

完整代码

#导入工具包
from scipy.spatial import distance as dist
from collections import OrderedDict
import numpy as np
import argparse
import time
import dlib
import cv2FACIAL_LANDMARKS_68_IDXS = OrderedDict([("mouth", (48, 68)),("right_eyebrow", (17, 22)),("left_eyebrow", (22, 27)),("right_eye", (36, 42)),("left_eye", (42, 48)),("nose", (27, 36)),("jaw", (0, 17))
])# http://vision.fe.uni-lj.si/cvww2016/proceedings/papers/05.pdf
def eye_aspect_ratio(eye):# 计算距离，竖直的A = dist.euclidean(eye[1], eye[5])B = dist.euclidean(eye[2], eye[4])# 计算距离，水平的C = dist.euclidean(eye[0], eye[3])# ear值ear = (A + B) / (2.0 * C)return ear# 输入参数
ap = argparse.ArgumentParser()
ap.add_argument("-p", "--shape-predictor",default="shape_predictor_68_face_landmarks.dat",help="path to facial landmark predictor")
ap.add_argument("-v", "--video", type=str, default="test.mp4",help="path to input video file")
args = vars(ap.parse_args())# 设置判断参数
EYE_AR_THRESH = 0.3
EYE_AR_CONSEC_FRAMES = 3# 初始化计数器
COUNTER = 0
TOTAL = 0# 检测与定位工具
print("[INFO] loading facial landmark predictor...")
detector = dlib.get_frontal_face_detector()
predictor = dlib.shape_predictor(args["shape_predictor"])# 分别取两个眼睛区域
(lStart, lEnd) = FACIAL_LANDMARKS_68_IDXS["left_eye"]
(rStart, rEnd) = FACIAL_LANDMARKS_68_IDXS["right_eye"]# 读取视频
print("[INFO] starting video stream thread...")
vs = cv2.VideoCapture(args["video"])
#vs = FileVideoStream(args["video"]).start()
time.sleep(1.0)def shape_to_np(shape, dtype="int"):# 创建68*2coords = np.zeros((shape.num_parts, 2), dtype=dtype)# 遍历每一个关键点# 得到坐标for i in range(0, shape.num_parts):coords[i] = (shape.part(i).x, shape.part(i).y)return coords# 遍历每一帧
while True:# 预处理frame = vs.read()[1]if frame is None:break(h, w) = frame.shape[:2]width=1200r = width / float(w)dim = (width, int(h * r))frame = cv2.resize(frame, dim, interpolation=cv2.INTER_AREA)gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)# 检测人脸rects = detector(gray, 0)# 遍历每一个检测到的人脸for rect in rects:# 获取坐标shape = predictor(gray, rect)shape = shape_to_np(shape)# 分别计算ear值leftEye = shape[lStart:lEnd]rightEye = shape[rStart:rEnd]leftEAR = eye_aspect_ratio(leftEye)rightEAR = eye_aspect_ratio(rightEye)# 算一个平均的ear = (leftEAR + rightEAR) / 2.0# 绘制眼睛区域leftEyeHull = cv2.convexHull(leftEye)rightEyeHull = cv2.convexHull(rightEye)cv2.drawContours(frame, [leftEyeHull], -1, (0, 255, 0), 1)cv2.drawContours(frame, [rightEyeHull], -1, (0, 255, 0), 1)# 检查是否满足阈值if ear < EYE_AR_THRESH:COUNTER += 1else:# 如果连续几帧都是闭眼的，总数算一次if COUNTER >= EYE_AR_CONSEC_FRAMES:TOTAL += 1# 重置COUNTER = 0# 显示cv2.putText(frame, "Blinks: {}".format(TOTAL), (10, 30),cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)cv2.putText(frame, "EAR: {:.2f}".format(ear), (300, 30),cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)cv2.imshow("Frame", frame)key = cv2.waitKey(10) & 0xFFif key == 27:breakvs.release()
cv2.destroyAllWindows()