Practical Computer Vision with SimpleCV

Chapter 3. Image Sources

This chapter examines various video and image sources in more detail. This is an important prerequisite to image processing. Obviously having a source of images is helpful before processing any images. The SimpleCV framework can capture images from a variety of image sources, ranging from a standard webcam to the Microsoft Kinect. In particular, this chapter covers:

• A review of working with webcams

• How to use a Kinect to capture depth information for basic 3D processing

• Using an IP camera as a digital video source

• Working with virtual devices to process images from video feeds or pre-captured sets of images

• How to handle a single image and a set of images

Overview

The SimpleCV framework supports most cameras that connect to a computer through a variety of interfaces such as USB, FireWire, or a built-in webcam. It can access networked IP cameras that are connected via a wire or a wireless connection. It can even interact with many video capture boards, which work with a variety of analog video inputs (such as Composite, Component, and S-Video).

Outside of connecting to the camera, there are options such as whether to use a monochrome or a color camera, or whether to use a camera with a CCD or a CMOS image sensor. Some cameras record different portions of the light spectrum, such as visible, infrared, or ultraviolet. Then there’s always the choice of how much to invest, as cameras these days range from the very inexpensive to the very expensive. With all of these ...