Welcome to MicroROS-Pi5 repository

Face recognition tracking

Note: The ROS_DOMAIN_ID of the Raspberry Pi and the microROS control board need to be consistent. You can check [MicroROS Control Board Parameter Configuration] to set the microROS control board ROS_DOMAIN_ID. Check the tutorial [Connect MicroROS Agent] to determine whether the IDs are consistent.

1、Program Description

After running the program, when a person's face is displayed on the screen and a box appears surrounding the person's face, the PTZ camera will move along with the movement of the person's face.

2、Steps

Program code reference path:

/root/yahboomcar_ws/src/yahboomcar_astra/yahboomcar_astra/face_fllow.py

2.1、Start command

After entering the docker container, according to the actual car model, enter the terminal.

xxxxxxxxxx
ros2 run yahboomcar_astra face_fllow  

After the program is started, the following camera screen will appear.

When the face is recognized, the frame will be selected, and the 2D pan/tilt will follow the movement of the face, and the terminal will print the movement angle.

2.2、Dynamic parameter adjustment

Enter the following commands in the Docker terminal.

xxxxxxxxxx
ros2 run rqt_reconfigure rqt_reconfigure

After modifying the parameters, click on a blank space in the GUI to write the parameter values. As can be seen from the above figure,

• faceTracker mainly adjusts the three parameters of PID to make the gimbal more sensitive

3、core code

3.1、face_fllow .py

The principle of function implementation is similar to that of object tracking. It calculates the rotation angle of the servo based on the center coordinates of the target, and then publishes it to the chassis. Part of the code is as follows.

x
# Calculate center point
 cv.putText(frame, text, (20, 30), cv.FONT_HERSHEY_SIMPLEX, 0.9, (0, 0, 255), 1)
        face_patterns = cv2.CascadeClassifier('/root/yahboomcar_ws/src/yahboomcar_astra/yahboomcar_astra/haarcascade_frontalface_default.xml')
        faces = face_patterns.detectMultiScale(frame , scaleFactor=1.1, minNeighbors=5, minSize=(100, 100))
        if len(faces)>0:
            for (x, y, w, h) in faces:
                m=x
                n=y
                cv2.rectangle(frame, (x, y), (x + w, y + h), (0, 255, 0), 2)
            self.execute(m,n)
#Calculate the steering gear rotation based on the center point
 [x_Pid, y_Pid] = self.PID_controller.update([point_x - 320, point_y - 240])
        if self.img_flip == True:
            self.PWMServo_X += x_Pid
            self.PWMServo_Y += y_Pid
        else:
            self.PWMServo_X  -= x_Pid
            self.PWMServo_Y  += y_Pid

        if self.PWMServo_X  >= 90:
            self.PWMServo_X  = 90
        elif self.PWMServo_X  <= -90:
            self.PWMServo_X  = -90
        if self.PWMServo_Y >= 20:
            self.PWMServo_Y = 20
        elif self.PWMServo_Y <= -90:
            self.PWMServo_Y = -90