Dekker1/ResearchMethods
Archived
1
0
Fork 0
Code

minor edits

Browse Source
This commit is contained in:
Silver-T 2018-05-25 19:47:21 +10:00
parent 81054d171b
commit 5cfa991e54
1 changed files with 13 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

27
mini_proj/report/waldo.tex
View File

@ -46,8 +46,7 @@
\end{abstract} \end{abstract}
\section{Introduction} \section{Introduction}
\tab Almost every child around the world knows about ``Where's Waldo?'', also
Almost every child around the world knows about ``Where's Waldo?'', also
known as ``Where's Wally?'' in some countries. This famous puzzle book has known as ``Where's Wally?'' in some countries. This famous puzzle book has
spread its way across the world and is published in more than 25 different spread its way across the world and is published in more than 25 different
languages. The idea behind the books is to find the character Waldo, languages. The idea behind the books is to find the character Waldo,
@ -59,7 +58,7 @@
that even adults will have trouble spotting Waldo is the fact that the that even adults will have trouble spotting Waldo is the fact that the
pictures are full of ``Red Herrings'': things that look like (or are colored pictures are full of ``Red Herrings'': things that look like (or are colored
as) Waldo, but are not actually Waldo. as) Waldo, but are not actually Waldo.
\\
\begin{figure}[ht] \begin{figure}[ht]
\includegraphics[scale=0.35]{waldo.png} \includegraphics[scale=0.35]{waldo.png}
\centering \centering
@ -69,7 +68,7 @@
} }
\label{fig:waldo} \label{fig:waldo}
\end{figure} \end{figure}
\par
The task of finding Waldo is something that relates to a lot of real life The task of finding Waldo is something that relates to a lot of real life
image recognition tasks. Fields like mining, astronomy, surveillance, image recognition tasks. Fields like mining, astronomy, surveillance,
radiology, and microbiology often have to analyse images (or scans) to find radiology, and microbiology often have to analyse images (or scans) to find
@ -77,7 +76,8 @@
are especially hard when the thing(s) you are looking for are similar to the are especially hard when the thing(s) you are looking for are similar to the
rest of the images. These tasks are thus generally performed using computers rest of the images. These tasks are thus generally performed using computers
to identify possible matches. to identify possible matches.
\\
\par
``Where's Waldo?'' offers us a great tool to study this kind of problem in a ``Where's Waldo?'' offers us a great tool to study this kind of problem in a
setting that is humanly tangible. In this report we will try to identify setting that is humanly tangible. In this report we will try to identify
Waldo in the puzzle images using different classification methods. Every Waldo in the puzzle images using different classification methods. Every
@ -93,7 +93,7 @@
\section{Background} \label{sec:background} \section{Background} \label{sec:background}
The classification methods used can separated into two separate groups: \tab The classification methods used can separated into two separate groups:
classical machine learning methods and neural network architectures. Many of classical machine learning methods and neural network architectures. Many of
the classical machine learning algorithms have variations and improvements the classical machine learning algorithms have variations and improvements
for various purposes; however, for this report we will be using their only for various purposes; however, for this report we will be using their only
@ -103,7 +103,7 @@
\subsection{Classical Machine Learning Methods} \subsection{Classical Machine Learning Methods}
The following paragraphs will give only brief descriptions of the different \tab The following paragraphs will give only brief descriptions of the different
classical machine learning methods used in this reports. For further reading classical machine learning methods used in this reports. For further reading
we recommend reading ``Supervised machine learning: A review of we recommend reading ``Supervised machine learning: A review of
classification techniques'' \cite{MLReview}. classification techniques'' \cite{MLReview}.
@ -158,8 +158,7 @@
of randomness and the mean of these trees is used which avoids this problem. of randomness and the mean of these trees is used which avoids this problem.
\subsection{Neural Network Architectures} \subsection{Neural Network Architectures}
\tab There are many well established architectures for Neural Networks depending
There are many well established architectures for Neural Networks depending
on the task being performed. In this paper, the focus is placed on on the task being performed. In this paper, the focus is placed on
convolution neural networks, which have been proven to effectively classify convolution neural networks, which have been proven to effectively classify
images \cite{NIPS2012_4824}. One of the pioneering works in the field, the images \cite{NIPS2012_4824}. One of the pioneering works in the field, the
@ -185,7 +184,7 @@
\section{Method} \label{sec:method} \section{Method} \label{sec:method}
In order to effectively utilize the aforementioned modeling and \tab In order to effectively utilize the aforementioned modeling and
classification techniques, a key consideration is the data they are acting classification techniques, a key consideration is the data they are acting
on. A dataset containing Waldo and non-Waldo images was obtained from an on. A dataset containing Waldo and non-Waldo images was obtained from an
Open Database\footnote{``The Open Database License (ODbL) is a license Open Database\footnote{``The Open Database License (ODbL) is a license
@ -238,7 +237,7 @@
\subsection{Neural Network Training}\label{nnTraining} \subsection{Neural Network Training}\label{nnTraining}
The neural networks used to classify the images were supervised learning \tab The neural networks used to classify the images were supervised learning
models; requiring training on a dataset of typical images. Each network was models; requiring training on a dataset of typical images. Each network was
trained using the preprocessed training dataset and labels for 25 epochs trained using the preprocessed training dataset and labels for 25 epochs
(one forward and backward pass of all data) in batches of 150. The number of (one forward and backward pass of all data) in batches of 150. The number of
@ -258,14 +257,14 @@
% Kelvin Start % Kelvin Start
\subsection{Benchmarking}\label{benchmarking} \subsection{Benchmarking}\label{benchmarking}
In order to benchmark the Neural Networks, the performance of these \tab In order to benchmark the Neural Networks, the performance of these
algorithms are evaluated against other Machine Learning algorithms. We use algorithms are evaluated against other Machine Learning algorithms. We use
Support Vector Machines, K-Nearest Neighbors (\(K=5\)), Naive Bayes and Support Vector Machines, K-Nearest Neighbors (\(K=5\)), Naive Bayes and
Random Forest classifiers, as provided in Scikit-Learn~\cite{scikit-learn}. Random Forest classifiers, as provided in Scikit-Learn~\cite{scikit-learn}.
\subsection{Performance Metrics}\label{performance-metrics} \subsection{Performance Metrics}\label{performance-metrics}
To evaluate the performance of the models, we record the time taken by \tab To evaluate the performance of the models, we record the time taken by
each model to train, based on the training data and the accuracy with which each model to train, based on the training data and the accuracy with which
the model makes predictions. We calculate accuracy as the model makes predictions. We calculate accuracy as
\[a = \frac{|correct\ predictions|}{|predictions|} = \frac{tp + tn}{tp + tn + fp + fn}\] \[a = \frac{|correct\ predictions|}{|predictions|} = \frac{tp + tn}{tp + tn + fp + fn}\]
@ -275,7 +274,7 @@
\section{Results} \label{sec:results} \section{Results} \label{sec:results}
The time taken to train each of the neural networks and traditional \tab The time taken to train each of the neural networks and traditional
approaches was measured and recorded alongside their accuracy (evaluated approaches was measured and recorded alongside their accuracy (evaluated
using a separate test dataset) in Table \ref{tab:results}. using a separate test dataset) in Table \ref{tab:results}.