Welcome to Jetson AGX Thor repository

Machine Code Recognition

Hardware Connection

Plug the USB camera into the motherboard's USB port. Open the system terminal and enter the following command to check the camera's device number. If /dev/video0 appears, it's normal.

ls /dev/video*

Install Driver Libraries

xxxxxxxxxx
sudo apt install libzbar-dev
sudo pip3 install pyzbar --break-system-packages
sudo pip3 install pupil-apriltags --break-system-packages

例程代码

​x
from pupil_apriltags import Detector
import cv2 as cv
from pyzbar.pyzbar import decode
import time
import numpy as np
from PIL import Image, ImageDraw, ImageFont
​
# Create a detector instance (parameters can be specified)
at_detector = Detector(
    families="tag36h11",  # Specify Tag Family
    nthreads=1,
    quad_decimate=1.0,
    quad_sigma=0.0,
    refine_edges=1,
    decode_sharpening=0.25,
    debug=0
)
​
​
def decodeDisplay(image):
    gray = cv.cvtColor(image, cv.COLOR_BGR2GRAY)
    # Perform Detection
    results = at_detector.detect(gray)
    for tag in results:
        # 提取机器码的边界框的位置
        # Extract the position of the boundary box of the apriltag
        (c1, c2, c3, c4) = tag.corners
        x = int(c4[0])
        y = int(c4[1])
        x1 = int(c2[0])
        y1 = int(c2[1])
        cv.rectangle(image, (x, y), (x1, y1), (225, 0, 0), 5)
​
        # 绘出图像上数据和类型
        # Draw the data and type on the image
        pilimg = Image.fromarray(image)
        # 创建画笔
        # create brush
        draw = ImageDraw.Draw(pilimg)
        # # 参数1：打印坐标，参数2：文本，参数3：字体颜色
        # Parameter 1: print coordinates, parameter 2: text, parameter 3: font color
        text = str(tag.tag_family).lstrip('b\'').rstrip('\'') + ' - ' + str(tag.tag_id)
        draw.text((x, y - 40), text, fill=(255, 0, 0))
        
        # 绘出标签码的中心点
        # Draw the center of the apriltag
        tag_center = tag.center
        radius = 5
        draw.ellipse([(tag_center[0]-radius, tag_center[1]-radius), (tag_center[0]+radius, tag_center[1]+radius)], fill="red")
        
        # # PIL图片转cv2 图片
        # PIL picture to CV2 picture
        image = cv.cvtColor(np.array(pilimg), cv.COLOR_RGB2BGR)
        # 向终端打印信息
        # Print information to terminal
        print(f"Tag Family: {tag.tag_family}")
        print(f"Tag ID: {tag.tag_id}")
        print(f"Center: {tag.center}")
        print(f"Corners: {tag.corners}")
        print("---")
    return image
​
if __name__ == '__main__':
    capture = cv.VideoCapture(0)
    cv_edition = cv.__version__
    if cv_edition[0] == '3': capture.set(cv.CAP_PROP_FOURCC, cv.VideoWriter_fourcc(*'XVID'))
    else: capture.set(cv.CAP_PROP_FOURCC, cv.VideoWriter.fourcc('M', 'J', 'P', 'G'))
    capture.set(cv.CAP_PROP_FRAME_WIDTH, 640)
    capture.set(cv.CAP_PROP_FRAME_HEIGHT, 480)
    print ("capture get FPS : ",capture.get(cv.CAP_PROP_FPS))
    while capture.isOpened():
        start = time.time()
        ret, frame = capture.read()
        action = cv.waitKey(10) & 0xFF
        frame = decodeDisplay(frame)
        end = time.time()
        fps = 1 / (end - start)
        text = "FPS : " + str(int(fps))
        cv.putText(frame, text, (30, 30), cv.FONT_HERSHEY_SIMPLEX, 0.6, (100, 200, 200), 1)
        cv.imshow('frame', frame)
        if action == ord('q') or action == 113: break
    capture.release()
    cv.destroyAllWindows()

Running the Program

Open a desktop terminal and execute the program:

xxxxxxxxxx
cd ~/content_recognition/apriltag/
python apriltag_parsing_usb.py

Note: Since the program requires a display window to show the video feed, do not run it via a remote SSH terminal. Please run the program using the native desktop environment or a VNC remote desktop session instead.

Program Output

Upon execution, the program automatically activates the USB camera. Simply point the camera at an AprilTag; the program will then automatically frame the tag in the video feed and print its associated information to the terminal.

Sample AprilTag Image: