Image Processing Reference
In-Depth Information
3
Basic image processing
operations
3.1
Overview
We shall now start to process digital images as described in Table 3.1. First, we shall
describe the brightness variation in an image using its histogram. We shall then look at
operations which manipulate the image so as to change the histogram, processes that shift
and scale the result (making the image brighter or dimmer, in different ways). We shall also
consider thresholding techniques that turn an image from grey level to binary. These are
called single point operations. After, we shall move to group operations where the group
is those points found inside a template. Some of the most common operations on the
groups of points are statistical, providing images where each point is the result of, say,
averaging the neighbourhood of each point in the original image. We shall see how the
statistical operations can reduce noise in the image, which is of benefit to the feature
extraction techniques to be considered later. As such, these basic operations are usually for
pre-processing for later feature extraction or to improve display quality.
3.2
Histograms
The intensity
histogram
shows how individual brightness levels are occupied in an image;
the
image contrast
is measured by the range of brightness levels. The histogram plots the
number of pixels with a particular brightness level against the brightness level. For 8-bit
pixels, the brightness ranges from zero (black) to 255 (white). Figure
3.1
shows an image
of an eye and its histogram. The histogram, Figure
3.1
(b), shows that not all the grey levels
are used and the lowest and highest intensity levels are close together, reflecting moderate
contrast. The histogram has a region between 100 and 120 brightness values which contains
the dark portions of the image, such as the hair (including the eyebrow) and the eye's iris.
The brighter points relate mainly to the skin. If the image was darker, overall, then the
histogram would be concentrated towards black. If the image was brighter, but with lower
contrast, then the histogram would be thinner and concentrated near the whiter brightness
levels.
This histogram shows us that we have not used all available grey levels. Accordingly, we
can stretch the image to use them all, and the image would become clearer. This is essentially
cosmetic attention to make the image's appearance better. Making the appearance better,
especially in view of later processing, is the focus of many basic image processing operations,
as will be covered in this chapter. The histogram can also reveal if there is noise in the
