# Introduction

Given images of rotated one-cent coins of varying (1) shininess/color, (2) years/mintmarks, and (3) levels of wear, as shown below, your objective is to detect the angle of rotation of each coin from the upright position.

There are two classes of coins: (1) the US Lincoln head one cent and (2) the US Lincoln memorial (tail) one cent.  It is expected that your algorithm can detect the rotation on both classes.

![Labeled Rotated Lincoln Heads](https://i.imgur.com/3ktjY8C.png)

![Labeled Rotated Memorial](https://i.imgur.com/sl6io3h.png)

# Dataset

The dataset consists of 568 images of rotated Lincoln heads and 532 images of Lincoln memorials.  All images are in color and have size 750 px by 750 px.  Each image is pre-cropped to only contain the circular coin; however, you may want to crop further to remove the coin edge.

![Rotated Dataset](https://i.imgur.com/6QFt3pM.png)