Image analysis and computer vision have always been important in industrial and scientific applications. With the popularization of cell phones with powerful cameras and internet connections, images are now increasingly generated by consumers. Therefore, there are opportunities to make use of computer vision to provide a better user experience in new contexts.

In this chapter, we will look at how to apply several techniques you have learned about in the rest of this book to this specific type of data. In particular, we will learn how to use the mahotas computer vision package to extract features from images. These features can then be used as input to the same classification methods we studied in other chapters. We will apply these techniques to publicly available datasets of photographs. We will also see how the same features can be used for finding similar images. We will also learn about using local features. These are relatively generic and achieve very good results in many tasks (although they have a higher computational cost).

Finally, at the end of the chapter, we will use Tensorflow to generate new images based on an existing dataset. In particular, in this chapter we will do the following:

• Learn how to represent and manipulate images as NumPy arrays
• Learn how to represent images as a small set of features so that standard classification and clustering methods can be used on this data type
• Learn how to use visual words to generate another type of feature
• Learn how to generate new images similar to existing ones