Published February 1, 2020

Gaussian Distribution?

Hello... did you ever want to set up a gaussian distribution on your BeagleBoard. org family of boards, e.g. the BBAI?

BeginnerProtip1 hour2,829

Things used in this project

Hardware components

BeagleBoard.org BeagleBone AI

Fan 5v

USB 3.0 Hub

keyboard (USB)

mouse (USB)

Micro HDMI Cable to HDMI Cable

Software apps and online services

Debian

Story

Hello,

So, I thought, "Why not learn something new with a new board?" This brought me to a specific spot in life where I can adapt to changing names, technology, and traditional scripting languages. Python!

So, let us start by putting the image on your AI board from BeagleBoard.org. If you go to their website, beagleboard.org, you will find a page dedicated to installing your image on your board if you do not want to create your own images.

This is handy in case you want to get up and running w/out various issues.

...

So, once you have your equipment, AI, and the image on the AI, we can start to use it and create a gaussian distribution like on the photo that is heading this story.

Oh and if you do not want to use your hardware, keyboard and mouse w/ the USB hub, just use tightvncserver on your AI w/ VNC Viewer on your Windows development desktop. If you have a Linux development desktop, you can use many tools outside of VNC Viewer to view your OS visually.

For example:

Those two packages are available on Linux along w/ a host of other applications.

Oh, so.

Where did I receive this source? Hmm? I do not know and I think since the source has not applicable notation to someone or some industry standard. I guess I will keep looking around until I find who this source belongs to currently.

Outside of that discussion, I will post the source at the end of this quick presentation.

...

So, get PuTTY if you are on WIndows Desktop or use your ssh client in Debian or Ubuntu if using Linux.

Okay, so.

We should have done some research and got around to getting familiar w/ the board by now, getting some ideas revolving around OpenCV, and gaussian distributions.

Now, since this is an easy example w/out files, we can just run it w/ python <YourFile.py> and that is that for now.

If you do not understand, enjoy reading more on this subject of AI, ML, and NN.

AI: Artificial Intelligence
ML: Machine Learning
NN: Neural Networks

Since we are just promoting a specific type of distribution or bell curve which does not represent our photo, look here for more ideas relating to this bell curve or normal/gaussian distribution.

https://www.khanacademy.org/math/statistics-probability/modeling-distributions-of-data/more-on-normal-distributions/v/introduction-to-the-normal-distribution

...

The fellow put a nice presentation together for getting started on ideas revolving around calculus, probability, and the bell curve along w/ two excel spreadsheets.

Enjoy.

Seth

P.S. Oh, if you do not have anything on your AI for some reason, python3-opencv and/or tightvncserver, use apt to install it.

Code

Bell Curve (Circles)?

#!/usr/bin/env python

# Python 2/3 compatibility
from __future__ import print_function
import sys
PY3 = sys.version_info[0] == 3

if PY3:
    xrange = range

import numpy as np
from numpy import random
import cv2

def make_gaussians(cluster_n, img_size):
    points = []
    ref_distrs = []
    for i in xrange(cluster_n):
        mean = (0.1 + 0.8*random.rand(2)) * img_size
        a = (random.rand(2, 2)-0.5)*img_size*0.1
        cov = np.dot(a.T, a) + img_size*0.05*np.eye(2)
        n = 100 + random.randint(900)
        pts = random.multivariate_normal(mean, cov, n)
        points.append( pts )
        ref_distrs.append( (mean, cov) )
    points = np.float32( np.vstack(points) )
    return points, ref_distrs

def draw_gaussain(img, mean, cov, color):
    x, y = np.int32(mean)
    w, u, vt = cv2.SVDecomp(cov)
    ang = np.arctan2(u[1, 0], u[0, 0])*(180/np.pi)
    s1, s2 = np.sqrt(w)*3.0
    cv2.ellipse(img, (x, y), (s1, s2), ang, 0, 360, color, 1, cv2.LINE_AA)


if __name__ == '__main__':
    cluster_n = 5
    img_size = 512

    print('press any key to update distributions, ESC - exit\n')

    while True:
        print('sampling distributions...')
        points, ref_distrs = make_gaussians(cluster_n, img_size)

        print('EM (opencv) ...')
        em = cv2.ml.EM_create()
        em.setClustersNumber(cluster_n)
        em.setCovarianceMatrixType(cv2.ml.EM_COV_MAT_GENERIC)
        em.trainEM(points)
        means = em.getMeans()
        covs = em.getCovs()  # Known bug: https://github.com/opencv/opencv/pull/4232
        found_distrs = zip(means, covs)
        print('ready!\n')
		
		img = np.zeros((img_size, img_size, 3), np.uint8)
        for x, y in np.int32(points):
            cv2.circle(img, (x, y), 1, (255, 255, 255), -1)
        for m, cov in ref_distrs:
            draw_gaussain(img, m, cov, (0, 255, 0))
        for m, cov in found_distrs:
            draw_gaussain(img, m, cov, (0, 0, 255))

        cv2.imshow('gaussian mixture', img)
        ch = cv2.waitKey(0)
        if ch == 27:
            break
    cv2.destroyAllWindows()

Credits

Seth

32 projects • 13 followers

Stay there and someone will find you...

Thanks to Beagleboard.org people for Producing the AI, OpenCV developers, and Unknown for now.

Gaussian Distribution?

Things used in this project

Hardware components

Software apps and online services

Story

Schematics

BeagleBoard.org Came Out w/ a New Board

Code

Bell Curve (Circles)?

Credits

Seth

Comments

Embed the widget on your own site

Gaussian Distribution?

Gaussian Distribution?

Things used in this project

Hardware components

Software apps and online services

Story

Schematics

BeagleBoard.org Came Out w/ a New Board

Code

Bell Curve (Circles)?

Credits

Seth

Comments

Related channels and tags