Published October 10, 2021

Bleeding Edge OpenCV and BBAI/Processing!

Oh! Hello... did you want OpenCV to work and finally get your edge. py file to work on call?

BeginnerProtip1 hour1,116

Bleeding Edge OpenCV and BBAI/Processing!

Things used in this project

Hardware components

BeagleBoard.org BeagleBone AI

Software apps and online services

Debian

Python3.9

OpenCV – Open Source Computer Vision Library OpenCV

Story

Hello,

W/ all of what is going on w/ BeagleBoard.org these days, the updated images I am using from ` https://rcn-ee.net/rootfs/debian-armhf/ ` can be mistook for something they are not. Use the Ethernet Port on your AI for this processing while the people at the .org try their magical potions of computer etiquette for WiFi, GPIO, and PRUs!

For the longest time, I have been trying to natively compile OpenCV (bleeding edge from their git repo) on the BBAI. Finally!

With Bullseye, I am able to do just that, "Compilation on the AI."

So, w/out further ado, here is a photo of the edge.py file in OpenCV samples/python/ one can build and change to suit their needs.

...

As you can tell, some of the edging is highlighted. Nice!

Here, after changing the file a bit, you can see another set of edging.

From neon green to yellow, one can alter the video in question to handle all sorts of nice, "fashionable" coloring.

...

So, now comes the part of natively compiling.

On the OpenCV website you can find this page: https://docs.opencv.org/master/d7/d9f/tutorial_linux_install.html and https://docs.opencv.org/master/d0/d76/tutorial_arm_crosscompile_with_cmake.html.

So, I went w/ the /tutorial_linux_install.html page first.

So, we would need to do a couple of steps to handle the native compilation.

# Download and unpack sources
wget -O opencv.zip https://github.com/opencv/opencv/archive/master.zip

wget -O opencv_contrib.zip https://github.com/opencv/opencv_contrib/archive/master.zip

unzip opencv.zip

unzip opencv_contrib.zip

# Create build directory and switch into it
mkdir -p build && cd build

# Configure
cmake -DOPENCV_EXTRA_MODULES_PATH=../opencv_contrib-master/modules ../opencv-master

# Build
cmake --build .

...

That "should" put, after many hours, OpenCV (bleeding edge) on your BBAI.

One could cross-compile w/ the other link.

Once the compilation is done, use these commands.

sudo apt install python3-opencv libopencv-dev

This way, you can use the simple yet effective:

import cv2 as cv : This is for your set up of personal files w/ OpenCV that is now fully integrated and can be utilized.

Now, you can use Python3.9, which is on Bullseye, and you can use C/C++ to handle your favorite files in OpenCV.

Seth

P.S. Be careful which files you run, i.e. as they can overload the ARM onboard the BBAI. Some will halt your board if not careful. If you are offloading to slave processors, please join in. I sure could use some examples or understandings. I will keep up research in time to see if offloading to a EVE or DSP is possible.

Code

Some of the Sample Code

#!/usr/bin/env python3

'''
This sample demonstrates Canny edge detection.

Usage:
  edge.py [<video source>]

  Trackbars control edge thresholds.

'''

# Python 2/3 compatibility
from __future__ import print_function

import cv2 as cv
import numpy as np

# relative module
import video

# built-in module
import sys


def main():
    try:
        fn = sys.argv[1]
    except:
        fn = 0

    def nothing(*arg):
        pass

    cv.namedWindow('edge')
    cv.createTrackbar('thrs1', 'edge', 2000, 5000, nothing)
    cv.createTrackbar('thrs2', 'edge', 4000, 5000, nothing)

    cap = video.create_capture(fn)
    while True:
        _flag, img = cap.read()
        gray = cv.cvtColor(img, cv.COLOR_BGR2GRAY)
        thrs1 = cv.getTrackbarPos('thrs1', 'edge')
        thrs2 = cv.getTrackbarPos('thrs2', 'edge')
        edge = cv.Canny(gray, thrs1, thrs2, apertureSize=5)
        vis = img.copy()
        vis = np.uint8(vis/2.)
        vis[edge != 0] = (0, 255, 255) # Here are your colors!
        cv.imshow('edge', vis)
        ch = cv.waitKey(5)
        if ch == 27:
            break

    print('Done')


if __name__ == '__main__':
    print(__doc__)
    main()
    cv.destroyAllWindows()