The HIPR package contains a large number of images which can be used as a general purpose image library for image processing experiments. This section catalogues the images in the library and describes each image briefly. The image file naming convention is described in the section on filename conventions.
The images may be divided into two main categories. Raw images are images that have had almost no processing done to them since they were captured. They are the kinds of input image that standard image sensors --- video cameras, scanners, etc. --- produce. For the HIPR image library, such images were obtained by a variety of methods. Some were scanned in from photographs, some were captured using video cameras, and some were obtained from standard image libraries on the Internet. The second category of image consists of processed images. These are simply raw images to which at least one stage of `image processing' has been applied.
These images are catalogued in two different ways, firstly by subject and type, and secondly as a straight alphabetical index of image filenames. Note that the alphabetical listing is ONLY available in the hypermedia version of HIPR.
The subject/type catalogue is useful if you know the sort of raw image you are looking for and want to see if HIPR has anything similar in its library. The top level of this catalogue is a list of image types and categories. Underneath these are listed the different raw images that belong to each category. Finally, underneath each of the raw image headings is listed the various processed images that have been produced, starting with that raw image. A brief description of the processing required to produce each image is also included.
The alphabetical catalogue is more simple, and consists of a straight alphabetical list of the filenames of all the images in the HIPR library. Each filename is in fact a hyperlink to the relevant explanatory section in the subject/type catalogue (this is why it is only available in the hypermedia version). It is useful if you know the filename of a particular image, and want to find out where it comes from and how it was produced.
If you have an Internet connection, then it is possible to obtain further images from public domain databases on the Internet. A good starting point for looking for images is the Computer Vision Home Page which at the time of writing can be found at:
Alternatively, collections of images on CD-ROM can be purchased for a variety of purposes from appropriate suppliers. Computer magazines are generally a good source of manufacturers' addresses.
©2003 R. Fisher, S. Perkins,
A. Walker and E. Wolfart.