Welcome to Transbot SE repository

6、Face Tracking

6.1、Face detection Introduction

In OpenCV, face detection in video only reads each frame of image from the camera, and then uses the static image detection method for detection. Face detection requires the classifier:

@ Face detector(default): haarcascade_frontalface_default.xml

@ Face detector(fast Harr): haarcascade_frontalface_alt2.xml

@ Face detector(Side view): haarcascade_profileface.xml

@ Eye detector(left eye): haarcascade_lefteye_2splits.xml

@ Eye detector(right eye): haarcascade_righteye_2splits.xml

@ Mouth detector: haarcascade_mcs_mouth.xml

@ Nose detector: haarcascade_mcs_nose.xml

@ Body detector: haarcascade_fullbody.xml

@ Face detector(fast LBP): lbpcascade_frontalface.xml

@ Only open eyes can be detected: haarcascade_eye.xml

@ Only person with glasses can be detected: haarcascade_eye_tree_eyeglasses.xml

@ https://github.com/opencv/opencv/tree/master/data Download classifier file link

haarcascade_profileface.xml is the cascading data of Haar. This xml can be obtained from this link https://github.com/opencv/opencv/blob/master/data/haarcascades/haarcascade_profileface.xml

Next, we can start face detection by face_cascade.detectMultiScale (). We need to convert the image into a grayscale, then, transfer each frame of the image obtained by the camera into .detectMultiScale ().

(Note: we need to ensure enter the correct location of haarcascade_profileface.xml correctly.)

OpenCV API function:

detectMultiScale(const Mat& image, vector& objects, double scaleFactor=1.1,int minNeighbors, int flag,cvSize)

Parameter analysis:

image --- Input grayscale image

objects --- The rectangular box vector set of the detected object

scaleFactor --- Each scale parameter in the image scale, the default value is 1.1.

minNeighbors --- Default is 3.

minNeighbors --- The default value of 3 indicates that there are at least 3 overlap detections, so we think that the face is exist.

minSize --- Target minimum size

maxSize --- Target maximum size

6.2. Operation steps

6.2.1. Start

jetson motherboard/Raspberry Pi 4B

Start the face recognition tutorial (robot side)

roslaunch transbot_bringup bringup.launch
rosrun transbot_facetracker face_follow.py

Raspberry Pi 5

Before running, please confirm that the large program has been permanently closed

Enter docker

Note: If there is a terminal that automatically starts docker, or there is a docker terminal that has been opened, you can directly enter the docker terminal to run the command, and there is no need to manually start docker

Start docker manually

xxxxxxxxxx
./run_docker.sh

xxxxxxxxxx
roslaunch transbot_bringup bringup.launch

Enter the same docker from multiple terminals

Keep the program of the previous docker terminal running and open a new terminal

Enter the following command

xxxxxxxxxx
docker ps

Enter the same docker and use the following 18870bc3dc00 to modify the ID displayed on the actual terminal.

xxxxxxxxxx
docker exec -it 18870bc3dc00 /bin/bash

xxxxxxxxxx
rosrun transbot_facetracker face_follow.py

6.2.2. PID adjustment

Dynamic parameter tuning

jetson motherboard/Raspberry Pi 4B

xxxxxxxxxx
rosrun rqt_reconfigure rqt_reconfigure

Raspberry Pi 5

Enter the same docker from multiple terminals

Keep the program of the previous docker terminal running and open a new terminal

Enter the following command

xxxxxxxxxx
docker ps
 KCF target tracking
HD camera robotic arm carrying