Chapter 19

Classifying Images


Handling images with Python

Obtaining features from an image

Using image information to detect features from faces

Performing image classification tasks on images of faces

Among the five senses, sight is certainly the most powerful in conveying knowledge and information derived from the world outside. Many people feel that the gift of sight helps children know about the different things and persons around them. In addition, humans receive and transmit knowledge across time by means of pictures, visual arts, and textual documents. This chapter helps you understand how Python can help your computer interact with images.

Because sight is so important and precious, it’s similarly invaluable for a machine learning algorithm because sight opens the algorithm to new capabilities. Most information today is available in digital form (text, music, photos, and videos), yet, being able to read visual information in a binary format doesn’t help you to understand it and to use it properly. In recent years, one of the more important uses of vision in machine learning is to classify images for all sorts of reasons. This chapter helps you discover techniques for obtaining image features for use in machine learning tasks.

For example, robots need to know which objects they should avoid and which objects they need to work with, yet without image classification, the task is impossible. Humans also rely on image classification to perform tasks such as ...

Get Machine Learning For Dummies now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.