## 
## Attaching package: 'dplyr'
## The following objects are masked from 'package:stats':
## 
##     filter, lag
## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

Introduction

Search engines like Google and Bing have tools that allow the user to choose specific types of images. Some types of images, such as line drawings or animated gifs, are trivial, as the search engine can simply check whether the picture is black-and-white, or whether the image is a .gif file. Other types of images, such as photographs and clip art, are more difficult to distinguish. In this project, I will attempt to create a model that can categorize a given image as either a stock photo or a clip art image.

Clip art images are typically vector drawings made using a computer, and are often heavily stylized. For the purposes of this project, all clip art images have been converted to .png files. Stock photos are professional photographs, often of objects or people in particular poses. They are usually sold for commercial use in presentations, papers, etc. and can be edited using image editing software such as Photoshop. Here is an example of each type of image: