OpenCV Color Tracking

In this chapter, we add some functions to control peripheral interfaces in OpenCV. For example, the camera's pan-tilt will move, and please keep your hands or other fragile objects away from its rotation radius.

Preparation

As the product will run the main demo by default, and the main demo will occupy the camera resources, in this case, this tutorial is not applicable. Please terminate the main demo or reboot the robot after disabling the auto-running of the main demo.

这里需要注意的是，由于机器人主程序中使用了多线程且由 crontab 配置开机自动运行，所以常规的 sudo killall python 的方法通常是不起作用的，所以我们这里介绍禁用主程序自动运行的方法。

如果你已经禁用了机器人主程序的开机自动运行，则不需要执行下面的结束主程序章节。

Terminate Main Demo

1. Click on "+" 点击上方本页面选项卡旁边的 “+”号，会打开一个新的名为 Launcher 的选项卡。

2. Click on "Terminal" in Other, and open the terminal window.

3. Input bash in the terminal window and press Enter.

4. Now you can use "Bash Shell" to control the robot.

5. Input the command: crontab -e

6. If you are asked what the editor to be used, you can input 1 and then press Enter, select to use "Nano".

7. Open "crontab" config file, and you can see:

@reboot ~/ugv_pt_rpi/ugv-env/bin/python ~/ugv_pt_rpi/app.py >> ~/ugv.log 2>&1
@reboot /bin/bash ~/ugv_pt_rpi/start_jupyter.sh >> ~/jupyter_log.log 2>&1

8. Add # in front of ……app.py >> …… to comment out this line.

# @reboot ~/ugv_pt_rpi/ugv-env/bin/python ~/ugv_pt_rpi/app.py >> ~/ugv.log 2>&1
@reboot /bin/bash ~/ugv_pt_rpi/start_jupyter.sh >> ~/jupyter_log.log 2>&1

9. On the terminal interface, press "Ctrl + X" to exit, and it will query Save modified buffer?, press Y and Enter to save the modification.

10. Reboot the device, and note that 重启设备，注意该过程会暂时关闭当前的 jupyter Lab，如果你上一步没有注释掉 ……start_jupyter.sh >>…… 这一行，那么当机器人重新开机后，你仍然可以正常使用 jupyter Lab (JupyterLab 与机器人主程序 app.py 是互相独立运行的)，可能需要重新刷新页面。

11. 这里需要注意一点，由于下位机持续通过串口与上位机通信，上位机在重启过程中有可能会由于串口电平的连续变化不能正常开机，拿上位机为树莓派的情况举例，重启时树莓派关机后不会再开机，红灯常亮绿灯不亮，此时可以关闭机器人电源开关，再打开，机器人就能够正常重启了。

12. Input the command to reboot: sudo reboot

13. 等待设备重启后（重启过程中树莓派的绿灯会闪烁，当绿灯闪烁频率降低或灭掉后即代表已经启动成功），刷新页面，继续该教程的剩余部分。

Demo

Directly run the following demo:

1. Choose the following demo:

2. Run it by Shift + Enter.

3. View the real-time video window:

4. Press STOP to stop the real-time video and release the camera resources.

如果运行时不能看到摄像头实时画面

需要点击上方的 Kernel - Shut down all kernels
关闭本章节选项卡，再次打开
点击 STOP 释放摄像头资源后重新运行代码块
Reboot the device

Run the Demo

在本章教程中，摄像头云台会转动，确保你的手或其它易碎物品远离摄像头云台的转动半径。

我们在例程中默认检测蓝色小球，确保画面背景中没有蓝色物体影响颜色识别功能，你也可以通过二次开发来更改检测颜色（HSV色彩空间）。

import matplotlib.pyplot as plt
import cv2
from picamera2 import Picamera2
import numpy as np
from IPython.display import display, Image
import ipywidgets as widgets
import threading

# Stop button
# ================
stopButton = widgets.ToggleButton(
    value=False,
    description='Stop',
    disabled=False,
    button_style='danger', # 'success', 'info', 'warning', 'danger' or ''
    tooltip='Description',
    icon='square' # (FontAwesome names without the `fa-` prefix)
)


def gimbal_track(fx, fy, gx, gy, iterate):
    global gimbal_x, gimbal_y
    distance = math.sqrt((fx - gx) ** 2 + (gy - fy) ** 2)
    gimbal_x += (gx - fx) * iterate
    gimbal_y += (fy - gy) * iterate
    if gimbal_x > 180:
        gimbal_x = 180
    elif gimbal_x < -180:
        gimbal_x = -180
    if gimbal_y > 90:
        gimbal_y = 90
    elif gimbal_y < -30:
        gimbal_y = -30
    gimbal_spd = int(distance * track_spd_rate)
    gimbal_acc = int(distance * track_acc_rate)
    if gimbal_acc < 1:
        gimbal_acc = 1
    if gimbal_spd < 1:
        gimbal_spd = 1
    base.base_json_ctrl({"T":self.CMD_GIMBAL,"X":gimbal_x,"Y":gimbal_y,"SPD":gimbal_spd,"ACC":gimbal_acc})
    return distance


# Display function
# ================
def view(button):
    picam2 = Picamera2()
    picam2.configure(picam2.create_video_configuration(main={"format": 'XRGB8888', "size": (640, 480)}))
    picam2.start()
    display_handle=display(None, display_id=True)

    color_upper = np.array([120, 255, 220])
    color_lower = np.array([ 90, 120,  90])
    min_radius = 12
    track_color_iterate = 0.023
    
    while True:
        frame = picam2.capture_array()
        # frame = cv2.flip(frame, 1) # if your camera reverses your image

        # uncomment this line if you are using USB camera
        # frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)

        img = cv2.cvtColor(frame, cv2.COLOR_RGB2BGR)
        blurred = cv2.GaussianBlur(img, (11, 11), 0)
        hsv = cv2.cvtColor(blurred, cv2.COLOR_BGR2HSV)
        mask = cv2.inRange(hsv, color_lower, color_upper)
        mask = cv2.erode(mask, None, iterations=5)
        mask = cv2.dilate(mask, None, iterations=5)

        cnts = cv2.findContours(mask.copy(), cv2.RETR_EXTERNAL,
            cv2.CHAIN_APPROX_SIMPLE)
        cnts = imutils.grab_contours(cnts)
        center = None

        height, width = img.shape[:2]
        center_x, center_y = width // 2, height // 2

        if len(cnts) > 0:
            # find the largest contour in the mask, then use
            # it to compute the minimum enclosing circle and
            # centroid
            c = max(cnts, key=cv2.contourArea)
            ((x, y), radius) = cv2.minEnclosingCircle(c)
            M = cv2.moments(c)
            center = (int(M["m10"] / M["m00"]), int(M["m01"] / M["m00"]))

            # only proceed if the radius meets a minimum size
            if radius > min_radius:
                distance = gimbal_track(center_x, center_y, center[0], center[1], track_color_iterate) #
                cv2.circle(overlay_buffer, (int(x), int(y)), int(radius), (128, 255, 255), 1)
        
        
        _, frame = cv2.imencode('.jpeg', frame)
        display_handle.update(Image(data=frame.tobytes()))
        if stopButton.value==True:
            picam2.close()
            display_handle.update(None)
            
            
# Run
# ================
display(stopButton)
thread = threading.Thread(target=view, args=(stopButton,))
thread.start()

Navigation menu

30 OpenCV Color Tracking

Contents