Nickiel/WGU-Capstone
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity
WGU-Capstone/WGU-Capstone-User-Guide.html

754 lines
26 KiB
HTML
Raw Permalink Blame History

This file contains invisible Unicode characters

This file contains invisible Unicode characters that are indistinguishable to humans but may be processed differently by a computer. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

<!DOCTYPE html>
<html xmlns="http://www.w3.org/1999/xhtml" lang xml:lang>
<head>
<meta charset="utf-8" />
<meta name="generator" content="pandoc" />
<meta name="viewport" content="width=device-width, initial-scale=1.0, user-scalable=yes" />
<title>WGU Capstone User Guide</title>
<style>
code{white-space: pre-wrap;}
span.smallcaps{font-variant: small-caps;}
div.columns{display: flex; gap: min(4vw, 1.5em);}
div.column{flex: auto; overflow-x: auto;}
div.hanging-indent{margin-left: 1.5em; text-indent: -1.5em;}
ul.task-list[class]{list-style: none;}
ul.task-list li input[type="checkbox"] {
font-size: inherit;
width: 0.8em;
margin: 0 0.8em 0.2em -1.6em;
vertical-align: middle;
}
.display.math{display: block; text-align: center; margin: 0.5rem auto;}
pre > code.sourceCode { white-space: pre; position: relative; }
pre > code.sourceCode > span { line-height: 1.25; }
pre > code.sourceCode > span:empty { height: 1.2em; }
.sourceCode { overflow: visible; }
code.sourceCode > span { color: inherit; text-decoration: inherit; }
div.sourceCode { margin: 1em 0; }
pre.sourceCode { margin: 0; }
@media screen {
div.sourceCode { overflow: auto; }
}
@media print {
pre > code.sourceCode { white-space: pre-wrap; }
pre > code.sourceCode > span { text-indent: -5em; padding-left: 5em; }
}
pre.numberSource code
{ counter-reset: source-line 0; }
pre.numberSource code > span
{ position: relative; left: -4em; counter-increment: source-line; }
pre.numberSource code > span > a:first-child::before
{ content: counter(source-line);
position: relative; left: -1em; text-align: right; vertical-align: baseline;
border: none; display: inline-block;
-webkit-touch-callout: none; -webkit-user-select: none;
-khtml-user-select: none; -moz-user-select: none;
-ms-user-select: none; user-select: none;
padding: 0 4px; width: 4em;
color: #aaaaaa;
}
pre.numberSource { margin-left: 3em; border-left: 1px solid #aaaaaa; padding-left: 4px; }
div.sourceCode
{ }
@media screen {
pre > code.sourceCode > span > a:first-child::before { text-decoration: underline; }
}
code span.al { color: #ff0000; font-weight: bold; }
code span.an { color: #60a0b0; font-weight: bold; font-style: italic; }
code span.at { color: #7d9029; }
code span.bn { color: #40a070; }
code span.bu { color: #008000; }
code span.cf { color: #007020; font-weight: bold; }
code span.ch { color: #4070a0; }
code span.cn { color: #880000; }
code span.co { color: #60a0b0; font-style: italic; }
code span.cv { color: #60a0b0; font-weight: bold; font-style: italic; }
code span.do { color: #ba2121; font-style: italic; }
code span.dt { color: #902000; }
code span.dv { color: #40a070; }
code span.er { color: #ff0000; font-weight: bold; }
code span.ex { }
code span.fl { color: #40a070; }
code span.fu { color: #06287e; }
code span.im { color: #008000; font-weight: bold; }
code span.in { color: #60a0b0; font-weight: bold; font-style: italic; }
code span.kw { color: #007020; font-weight: bold; }
code span.op { color: #666666; }
code span.ot { color: #007020; }
code span.pp { color: #bc7a00; }
code span.sc { color: #4070a0; }
code span.ss { color: #bb6688; }
code span.st { color: #4070a0; }
code span.va { color: #19177c; }
code span.vs { color: #4070a0; }
code span.wa { color: #60a0b0; font-weight: bold; font-style: italic; }
</style>
<style type="text/css">html{
font-family:sans-serif;
-ms-text-size-adjust:100%;
-webkit-text-size-adjust:100%
}
body{
background-color: #11111b;
color: #cdd6f4;
margin:0;
}
a{
color: #a6d189;
background:transparent
}
a:active,a:hover{
outline:0
}
b,strong{
font-weight:bold
}
h1{
color: #eba0ac;
font-size:2em;
margin:0.67em 0
}
small{
font-size:80%
}
sub,sup{
color: #b5bfe2
font-size:75%;
line-height:0;
position:relative;
vertical-align:baseline
}
sup{
top:-0.5em
}
sub{
bottom:-0.25em
}
img{
border:0
}
hr{
-moz-box-sizing:content-box;
box-sizing:content-box;
height:0;
}
pre{
overflow:auto
}
code,kbd,pre,samp{
font-family:monospace, monospace;
font-size:1em;
color: #414559;
}
textarea{
overflow:auto
}
table{
border-collapse:collapse;
border-spacing:0
}
td,th{
padding:0
}
body,code,tr.odd,tr.even,body{
line-height:1.3;
text-align:justify;
-moz-hyphens:auto;
-ms-hyphens:auto;
-webkit-hyphens:auto;
hyphens:auto
}
@media (max-width: 400px){
body{
font-size:12px;
margin-left:10px;
margin-right:10px;
margin-top:10px;
margin-bottom:15px
}
}
@media (min-width: 401px) and (max-width: 600px){
body{
font-size:14px;
margin-left:10px;
margin-right:10px;
margin-top:10px;
margin-bottom:15px
}
}
@media (min-width: 601px) and (max-width: 900px){
body{
font-size:15px;
margin-left:100px;
margin-right:100px;
margin-top:20px;
margin-bottom:25px
}
}
@media (min-width: 901px) and (max-width: 1800px){
body{
font-size:17px;
margin-left:200px;
margin-right:200px;
margin-top:30px;
margin-bottom:25px;
max-width:800px
}
}
@media (min-width: 1801px){
body{
font-size:18px;
margin-left:20%;
margin-right:20%;
margin-top:30px;
margin-bottom:25px;
max-width:1000px
}
}
p{
margin-top:10px;
margin-bottom:18px
}
em{
font-style:italic
}
strong{
font-weight:bold
}
h1,h2,h3,h4,h5,h6{
font-weight:bold;
padding-top:0.25em;
margin-bottom:0.15em
}
header{
line-height:2.475em;
padding-bottom:0.7em;
border-bottom:1px solid #bbb;
margin-bottom:1.2em
}
header>h1{
border:none;
padding:0;
margin:0;
font-size:225%
}
header>h2{
border:none;
padding:0;
margin:0;
font-style:normal;
font-size:175%
}
header>h3{
padding:0;
margin:0;
font-size:125%;
font-style:italic
}
header+h1{
border-top:none;
padding-top:0px
}
h1{
border-top:1px solid #bbb;
padding-top:15px;
font-size:150%;
margin-bottom:10px
}
h1:first-of-type{
border:none
}
h2{
font-size:125%;
font-style:italic
}
h3{
font-size:105%;
font-style:italic
}
hr{
border:0px;
border-top:1px solid #bbb;
width:100%;
height:0px
}
hr+h1{
border-top:none;
padding-top:0px
}
ul,ol{
font-size:90%;
margin-top:10px;
margin-bottom:15px;
padding-left:30px
}
ul{
list-style:circle
}
ol{
list-style:decimal
}
ul ul,ol ol,ul ol,ol ul{
font-size:inherit
}
li{
margin-top:5px;
margin-bottom:7px
}
q,blockquote,dd{
font-style:italic;
font-size:90%
}
blockquote,dd{
quotes:none;
border-left:0.35em #bbb solid;
padding-left:1.15em;
margin:0 1.5em 0 0
}
blockquote blockquote,dd blockquote,blockquote dd,dd dd,ol blockquote,ol dd,ul blockquote,ul dd,blockquote ol,dd ol,blockquote ul,dd ul{
font-size:inherit
}
a,a:link,a:visited,a:hover{
text-decoration:none;
border-bottom:1px dashed #111
}
a:hover,a:link:hover,a:visited:hover,a:hover:hover{
border-bottom-style:solid
}
a.footnoteRef,a:link.footnoteRef,a:visited.footnoteRef,a:hover.footnoteRef{
border-bottom:none;
color:#666
}
code{
font-family:"Source Code Pro","Consolas","Monaco",monospace;
font-size:85%;
background-color:#ddd;
border:1px solid #bbb;
padding:0px 0.15em 0px 0.15em;
-webkit-border-radius:3px;
-moz-border-radius:3px;
border-radius:3px
}
pre{
margin-right:1.5em;
display:block
}
pre>code{
display:block;
font-size:70%;
padding:10px;
-webkit-border-radius:5px;
-moz-border-radius:5px;
border-radius:5px;
overflow-x:auto
}
blockquote pre,dd pre,ul pre,ol pre{
margin-left:0;
margin-right:0
}
blockquote pre>code,dd pre>code,ul pre>code,ol pre>code{
font-size:77.77778%
}
caption,figcaption{
font-size:80%;
font-style:italic;
text-align:right;
margin-bottom:5px
}
caption:empty,figcaption:empty{
display:none
}
table{
width:100%;
margin-top:1em;
margin-bottom:1em
}
table+h1{
border-top:none
}
tr td,tr th{
padding:0.2em 0.7em
}
tr.header{
border-top:1px solid #222;
border-bottom:1px solid #222;
font-weight:700
}
tr.odd{
background-color:#eee
}
tr.even{
background-color:#ccc
}
tbody:last-child{
border-bottom:1px solid #222
}
dt{
font-weight:700
}
dt:after{
font-weight:normal;
content:":"
}
dd{
margin-bottom:10px
}
img{
display:block;
margin:0px auto;
padding:0px;
max-width:100%
}
figcaption{
margin:5px 10px 5px 30px
}
.footnotes{
color:#666;
font-size:70%;
font-style:italic
}
.footnotes li p:last-child a:last-child{
border-bottom:none
}
</style>
<!--[if lt IE 9]>
<script src="//cdnjs.cloudflare.com/ajax/libs/html5shiv/3.7.3/html5shiv-printshiv.min.js"></script>
<![endif]-->
</head>
<body>
<nav id="TOC" role="doc-toc">
<ul>
<li><a href="#step-0-clone-the-repository" id="toc-step-0-clone-the-repository">Step 0: Clone the
repository</a></li>
<li><a href="#project-structure" id="toc-project-structure">Project
Structure</a>
<ul>
<li><a href="#top-level" id="toc-top-level">Top Level</a></li>
<li><a href="#cascades" id="toc-cascades">./Cascades</a></li>
<li><a href="#traning_data" id="toc-traning_data">./Traning_data</a></li>
<li><a href="#validation" id="toc-validation">./Validation</a></li>
</ul></li>
<li><a href="#step-1---prerequisites" id="toc-step-1---prerequisites">Step 1 - Prerequisites</a>
<ul>
<li><a href="#set-up-virtual-environment" id="toc-set-up-virtual-environment">Set up virtual environment</a></li>
<li><a href="#install-requirements" id="toc-install-requirements">Install requirements</a></li>
</ul></li>
<li><a href="#step-2---running-the-project" id="toc-step-2---running-the-project">Step 2 - Running the
project</a></li>
<li><a href="#additional-flags" id="toc-additional-flags">Additional
flags</a>
<ul>
<li><a href="#help" id="toc-help">Help</a></li>
<li><a href="#version" id="toc-version">Version</a></li>
<li><a href="#show-dashboard" id="toc-show-dashboard">Show
Dashboard</a></li>
<li><a href="#output-adjustment-instructions" id="toc-output-adjustment-instructions">Output Adjustment
Instructions</a></li>
<li><a href="#use-video-file" id="toc-use-video-file">Use Video
File</a></li>
<li><a href="#headless-mode" id="toc-headless-mode">Headless
Mode</a></li>
<li><a href="#save-frames-for-training-data" id="toc-save-frames-for-training-data">Save Frames for Training
Data</a></li>
<li><a href="#generate-validation-file" id="toc-generate-validation-file">Generate Validation File</a></li>
</ul></li>
<li><a href="#training-your-own-haar-file" id="toc-training-your-own-haar-file">Training Your Own Haar File</a>
<ul>
<li><a href="#prerequisites" id="toc-prerequisites">Prerequisites</a></li>
<li><a href="#generating-positive-images" id="toc-generating-positive-images">Generating positive images</a></li>
</ul></li>
<li><a href="#validation-and-testing" id="toc-validation-and-testing">Validation and Testing</a>
<ul>
<li><a href="#generate-the-ground-truth-file" id="toc-generate-the-ground-truth-file">Generate the Ground Truth
file</a></li>
<li><a href="#getting-the-model-validation-file" id="toc-getting-the-model-validation-file">Getting the model validation
file</a></li>
<li><a href="#comparing-it-to-the-ground-truth" id="toc-comparing-it-to-the-ground-truth">Comparing it to the ground
truth</a></li>
</ul></li>
</ul>
</nav>
<h1 id="step-0-clone-the-repository">Step 0: Clone the repository</h1>
<p>Before you can run this project, you will need to clone the git
repository with the following command:</p>
<div class="sourceCode" id="cb1"><pre class="sourceCode sh"><code class="sourceCode bash"><span id="cb1-1"><a href="#cb1-1" aria-hidden="true" tabindex="-1"></a><span class="fu">git</span> clone https://git.nickiel.net/Nickiel/WGU-Capstone</span></code></pre></div>
<p>See <a href="#project-structure">Project Structure</a> for more
information on the repository you just cloned.</p>
<p>See <a href="#step-1---prerequisites">Step 1 - Prerequisites</a> on
what is required before you can run this project.</p>
<h1 id="project-structure">Project Structure</h1>
<p>Below you can find the default project folder structure after cloning
it:</p>
<pre><code>WGU-Capstone
├.gitignore
├Main.py
├README.md
├WGU-Capstone-User-Guide.html
├requirements.txt
├shell.nix
├cascades
│ ├ cascade_1.xml
│ ├ cascade_2.xml
│ ├ cascade_5.xml
│ └ cascade_10.xml
├training_data
│ ├ positives
│ └ training_data_setup.py
└validation
├ TestVideo.mp4
├ compare_to_gt.py
├ create_ground_truth.py
└ ground_truth.txt</code></pre>
<p><a href="#step-1---prerequisites">Click here to skip the detailed
file structure explaination</a></p>
<h2 id="top-level">Top Level</h2>
<p>In the top-level of the cloned repository, you will find most of the
files required for the core-fuctionality.</p>
<h4 id="gitignore">.gitignore</h4>
<p>This file excludes files we don’t want to check into git - such as
the training data. These files continue to exist on machine, but they
are not uploaded to the remote git repository. This is very helpful to
keep the clone sizes down, and upload/download speeds up.</p>
<h4 id="main.py">Main.py</h4>
<p>The main file of interest for this project. This file contains all of
the code for the finished product. As long as this file is in the same
folder as the <code>./cascades</code> folder, it can be copied and run
anywhere with the prerequisites installed.</p>
<h4 id="readme.md">README.md</h4>
<p>The file you are reading in a format that most git hosting servers
automatically render as the home-page.</p>
<h4 id="wgu-capstone-user-guide.html">WGU-Capstone-User-Guide.html</h4>
<p>The html version of the README.md file that was bundled with CSS and
hyper-links.</p>
<h4 id="requirements.txt">requirements.txt</h4>
<p>The file that contains all of the python pip requiremnts to run. This
packages in this file can either be installed by hand
(e.g. <code>pip install opencv-python</code>), or can all be installed
at once with <code>pip install -r requirements.txt</code> which will
install all of the modules needed to run this project that are not
included in the standard library.</p>
<h4 id="shell.nix">shell.nix</h4>
<p>A file that can be used on <a href="https://nixos.org/">Nix and
NixOS</a> systems to create a reproducable environement with all of the
requirements to run the <code>Main.py</code> file.</p>
<h2 id="cascades">./Cascades</h2>
<p>This folder contains the final trained models created by this project
in the model training step. For more information on how they were
created, see <a href="#training-your-own-haar-file">Training your own
Haar file</a> below.</p>
<p>This folder needs to be in the same directory as the
<code>Main.py</code> file for the <code>Main.py</code> file to be able
to run.</p>
<h2 id="traning_data">./Traning_data</h2>
<p>This folder contains all of the requirements for creating a new model
from a few un-catagorized positive images, and a large dataset of
negatives.</p>
<p>NOTE: Before anything in this folder can be run, please see <a href="#training-your-own-haar-file">the section on training the haar
files</a> for several prerequisites.</p>
<h4 id="training_data_setup.py">./Training_data_setup.py</h4>
<p>This python file takes a large data-set of negative images from the
<code>./training_data/negatives</code> folder and creates .vec files
that can be passed as an arguement to the utility that trains the final
Haar file.</p>
<h4 id="positives">./Positives</h4>
<p>This folder contains the 10 images that were used to create the
cascade files included in this project. These files were included
because the 10 images are a very small dataset in comparison to the
required negatives.</p>
<h2 id="validation">./Validation</h2>
<p>The folder contains all of the scripts and files used to measure the
performance and accuracy of the generated models.</p>
<h4 id="testvideo.mp4">TestVideo.mp4</h4>
<p>This minute-long video was used to test the trained models.</p>
<h4 id="compare_to_gt.py">Compare_to_gt.py</h4>
<p>This file compares the output of a <code>--validate</code> output
file generated by <code>Main.py</code> of a run with the provided
<code>ground_truth.txt</code> file. The output of this file is a .csv
file that describes the average deviation from the boxes described by
the <code>ground_truth.txt</code> file. See <a href="#validation-and-testing">Validation and Testing</a> for more
information on this process.</p>
<h4 id="create_ground_truth.py">Create_ground_truth.py</h4>
<p>This is the file used to create the <code>ground_truth.txt</code>
file from the provided <code>TestVideo.mp4</code>.</p>
<h1 id="step-1---prerequisites">Step 1 - Prerequisites</h1>
<p>Before you can run this project, you need a python environment with
the required packages installed.</p>
<p>If you are using Nix or NixOS, simply run <code>nix shell</code> in
the <code>WGU-Capstone</code> folder, and all of the packages required
to run <code>Main.py</code> will be installed for that shell
session.</p>
<p>However, if you are not on a Nix system, continue reading.</p>
<p>The steps below detail how to set up a virtual environment that can
be used to run this project, but a system-wide install of python with
the packages detailed in <code>requirements.txt</code> installed will
also suffice.</p>
<h3 id="set-up-virtual-environment">Set up virtual environment</h3>
<p>This project was created with python 3.11, and other versions are not
garunteed to work. So to ensure the project works as designed, install
python 3.11 from the official python download page.</p>
<p>Once you have python 3.11 installed on your system, navigate to the
cloned repository’s root directory, and run the following command to
create a new virtual environement:</p>
<div class="sourceCode" id="cb3"><pre class="sourceCode python"><code class="sourceCode python"><span id="cb3-1"><a href="#cb3-1" aria-hidden="true" tabindex="-1"></a>python <span class="op">-</span>m venv .<span class="op">/</span>venv</span></code></pre></div>
<p>You can now run the following commands to enter the virtual
environment, and any python commands will be run inside the virtual
environment instead of your system-wide installation.</p>
<p>On windows run the following if you are using a cmd prompt:</p>
<pre class="shell"><code>.\venv\Scripts\activate.bat</code></pre>
<p>On windows in powershell:</p>
<pre class="shell"><code>.\venv\Scripts\Activate.ps1</code></pre>
<p>If you are on a linux based operating system, enter the virtual
environment with:</p>
<pre class="shell"><code>.\venv\Scripts\activate</code></pre>
<h3 id="install-requirements">Install requirements</h3>
<p>Now that you have activated the virtual environment, install the
non-standard library requirements with the below command:</p>
<pre class="shell"><code>pip install -r ./requirements.txt</code></pre>
<h1 id="step-2---running-the-project">Step 2 - Running the project</h1>
<p>Now that the pre-requisites have been installed, you can run the
project. For a full list of command-line arguements, run
<code>python Main.py --help</code>.</p>
<p>Run the project with the dashboard enabled with the following command
from the root of the project directory:</p>
<pre class="shell"><code>python Main.py -d</code></pre>
<p>You should see the web-cam of your computer turn on, and a window
appear showing the view of the webcam, with boxes around any detected
faces.</p>
<p>To display the calculated adjustment amounts generated by this
project, enable the print-to-stoud feature with the <code>-o</code>
flag:</p>
<pre class="shell"><code>python Main.py -d -o</code></pre>
<p>This command will output the calculated output commands for every
detected face, and also show the summary statistics.</p>
<h1 id="additional-flags">Additional flags</h1>
<p>This section will describe, in greater depth, the available feature
flags shown by the <code>--help</code> screen.</p>
<h2 id="help">Help</h2>
<p><code>-h</code> or <code>--help</code></p>
<p>Displays all of the available parameters and a quick description</p>
<h2 id="version">Version</h2>
<p><code>-v</code> or <code>--version</code></p>
<p>Prints the version of the program and exits</p>
<h2 id="show-dashboard">Show Dashboard</h2>
<p><code>-d</code> or <code>--dasbboard</code></p>
<p>Display the run-summary statistics, these are off by default.</p>
<h2 id="output-adjustment-instructions">Output Adjustment
Instructions</h2>
<p><code>-o</code> or <code>--output</code></p>
<p>Print the calculated adjustment instructions generated by the
program. This output demonstrates the generated values that will be sent
to the motorized camera platform.</p>
<h2 id="use-video-file">Use Video File</h2>
<p><code>-f &lt;file_path&gt;</code> or
<code>--file &lt;file_path&gt;</code></p>
<p>Use a video file (such as ./validation/TestVideo.mp4) instead of the
computer’s webcam. Useful for generating validation files and on
machines without a working webcam.</p>
<h2 id="headless-mode">Headless Mode</h2>
<p><code>-s</code> or <code>--no-screen</code></p>
<p>Run the program without the window displaying processed video
frames.</p>
<h2 id="save-frames-for-training-data">Save Frames for Training
Data</h2>
<p><code>-t</code> or <code>--training-data</code></p>
<p>Save frames where faces were found to <code>./output</code> as .jpg
files, and save the located face’s location to a .csv file. This feature
will be used to generate positive images automatically for training
future models.</p>
<h2 id="generate-validation-file">Generate Validation File</h2>
<p><code>--validate</code></p>
<p>Outputs all discovered boxes, the frame they were found on, and the
box coordinates so the model can be validated against the ground truth.
See <a href="#validation-and-testing">validation and testing</a> for
more information on this process.</p>
<h1 id="training-your-own-haar-file">Training Your Own Haar File</h1>
<p>This project contains the scripts required to train your own Haar
cascade files, but it does not contain several of the dependencies.</p>
<p>NOTE: These steps only apply to Windows devices.</p>
<h2 id="prerequisites">Prerequisites</h2>
<p>The first requirement needed before you can train your own Haar file,
is a large number of negative images. For this project, I used <a href="https://www.kaggle.com/datasets/arnaud58/landscape-pictures/">this
Kaggle dataset of landscape images</a> as my negatives datasource. After
downloading this file, unzip it and deposit all of the raw images into
the <code>./training_data/negatives</code> folder - create it if
needed.</p>
<p>Next we need to download the windows OpenCV binary distributable and
put it in our training_data folder.</p>
<p>You can download the 3.4.15 binary executable <a href="https://sourceforge.net/projects/opencvlibrary/files/3.4.15/opencv-3.4.15-vc14_vc15.exe/download">here</a>.
(You can also go <a href="https://opencv.org/releases/">here</a> and
find the 3.4.15 release and choose “windows” to get to the same
page).</p>
<p>After the .exe file has downloaded, open it and go through the steps
to unzip it. After it has been unzipped, copy the folder to
<code>./training_data/opencv</code>. So you should be able to run this
from the training_data directory:</p>
<pre class="shell"><code>.\opencv\build\x64\vc15\bin\opencv_createsamples.exe</code></pre>
<p>If you do not get an error running the above command, then it was
installed correctly.</p>
<h2 id="generating-positive-images">Generating positive images</h2>
<p>Now that we have the create_samples utility provided by OpenCV (they
stopped distributing executables of it after 3.4.15) and the negatives
folder full of negative images, we can use the
<code>training_data_setup.py</code> file to create several different
sized datasets ready for training Haar cascade files on.</p>
<p>The python file will run the create_samples tool for every positive
image in <code>./positives</code>, creating many positive images. The
script will do all of the steps up through creating the .vec files that
the train_cascade exectuable requires.</p>
<p>Before exiting, training_data_setup outputs the commands that need to
be run to train the models. Run these commands from the training_data
folder, and after they have finished training, you can use the generated
Haar cascades instead of the ones provided.</p>
<h1 id="validation-and-testing">Validation and Testing</h1>
<p>The following describes the process I used to test the precision and
accuracy of the generated cascade files.</p>
<h2 id="generate-the-ground-truth-file">Generate the Ground Truth
file</h2>
<p>I have included a generated <code>ground_truth.txt</code> file, so
you don’t need to do this step. But if you would like to generate the
ground truth file from the provided test video, navigate to root of the
project, and run the create ground truth script:</p>
<pre class="shell"><code>python create_ground_truth.py</code></pre>
<p>A window will open and display the process as it creates the file.
This script does not utilize Haar files, but the MIL tracking algorithm,
which results in much more accurate results, but a slower processing
speed for the video.</p>
<p>All of these settings have been hard-coded so it will always output
the same ground truth file.</p>
<h2 id="getting-the-model-validation-file">Getting the model validation
file</h2>
<p>Now that we have the ground truth for our Test Video, we need to
generate the same file with our trained model.</p>
<p>To do this, edit the <code>Main.py</code> file so that it uses the
new cascade, then run the python file with the <code>--validate</code>
option set, and the test video passed to the <code>-f</code> flag. The
command used to generate the statistics with the test video provided is
this:</p>
<pre class="shell"><code>python ./Main.py -d -f ./validation/TestVideo.mp4 --validate</code></pre>
<p>(Notice that we can still display the dashboard while it outputs
validation info)</p>
<p>This will create a new file in the <code>./validation</code> folder
describing the faces and locations found in each frame.</p>
<h2 id="comparing-it-to-the-ground-truth">Comparing it to the ground
truth</h2>
<p>I have created a script to automatically compare a validation file
with a ground truth file, and output the average absolute deviation in
adjustment instructions. It requires two arguements, and has one
optional output. You can see the options with the <code>--help</code>
flag, but I will demonstrate all of the options below.</p>
<p>You can use <code>./validation/compare_to_gt.py</code> like this:</p>
<pre class="shell"><code>cd ./validation
python compare_to_gt ./ground_truth.txt ./20231012-081235.txt ./output.csv --faces_count_file ./faces_count_output.csv</code></pre>
<p>This script will then take the generated test validation file, and
get what the generated adjustment output would be, and gets the absolute
difference between it and the ground truth, then it adds together all
results for each frame - this last part penalizes false positives. We
can then take the generated output file, and open it in Excel. We can
take the average of it to see what the average deviation from the ground
truth would be. The generated faces_count_output file contains the
number of faces found in each frame, and can be used to measure the
number of false positives.</p>
</body>
</html>
Reference in a new issue View git blame Copy permalink