Add chapter "Problems with image recognition" (#17)

IDPA16 · Nov 30, 2016 · 8502f2d · 8502f2d
1 parent 15e196f
commit 8502f2d
Show file tree

Hide file tree

Showing 5 changed files with 10 additions and 4 deletions.
diff --git a/main.tex b/main.tex
@@ -56,13 +56,13 @@
 
 	\section{Convolutional Neural Networks}
 		\subsection{Problems with image recognition}
-			.
+			\subfile{sections/"convolutional neural networks"/"problems with image recognition"}
 		\subsection{Subsampling}
-			\subfile{sections/"Convolutional Neural Networks/Subsampling/subsampling.tex"}
+			\subfile{sections/"convolutional neural networks/subsampling/subsampling.tex"}
 			\subsubsection{Kernels}
-			\subfile{sections/"Convolutional Neural Networks/Subsampling/kernerls.tex"}
+			\subfile{sections/"convolutional neural networks/subsampling/kernerls.tex"}
 			\subsubsection{Poolers}
-			\subfile{sections/"Convolutional Neural Networks/Subsampling/poolers.tex"}
+			\subfile{sections/"convolutional neural networks/subsampling/poolers.tex"}
 		\subsection{Structure}
 			.
 	\newpage

diff --git a/sections/convolutional neural networks/problems with image recognition.tex b/sections/convolutional neural networks/problems with image recognition.tex
@@ -0,0 +1,6 @@
+Most neural networks are unable to handle the amount of data contained in an image.
+For example an image with a resolution of 3264x2448 (8 Megapixels) would result in almost 24 million inputs, as each pixel is split into its red, green and blue parts.
+
+Another challenge is the detection of so called "features" across an image. Traditional neural networks only detect a feature at a specific location in the image.
+This is a big issue in image recognition, as you almost always want the entire image to be handled equally.
+A self-driving car should recognize a stop sign, regardless of its position in the image.
diff --git a/... Neural Networks/Subsampling/kernerls.tex → ... neural networks/subsampling/kernerls.tex b/... Neural Networks/Subsampling/kernerls.tex → ... neural networks/subsampling/kernerls.tex
diff --git a/...l Neural Networks/Subsampling/poolers.tex → ...l neural networks/subsampling/poolers.tex b/...l Neural Networks/Subsampling/poolers.tex → ...l neural networks/subsampling/poolers.tex
diff --git a/...ural Networks/Subsampling/subsampling.tex → ...ural networks/subsampling/subsampling.tex b/...ural Networks/Subsampling/subsampling.tex → ...ural networks/subsampling/subsampling.tex