4' image enhancement

1
4. ????(image enhancement)
? Introduction
- The objective of image enhancement is to
process an image so that the processed image is
more suitable than the original image for a
specific application. cf) In image restoration,
an ideal image has been degraded, and the
objective is to make the processed image resemble
the original image as much as possible. -
Classification of enhancement techniques 1)
point operations - contrast manipulation,
histogram modification 2) spatial operations -
noise smoothing, median filtering, edge
sharpening 3) transform operations -
homomorphic filtering
2
? Point operations
- Contrast Stretching Low-contrast images occur
often due to poor or non-uniform illumination or
due to non-linearity or small dynamic range of
the imaging sensor. The idea behind contrast
stretching is to increase the dynamic range of
the graylevels in the image being processed. The
slope of the transformation function is chosen
greater than unity in the region of the stretch.
3
Contrast stretching transformation function
- Clipping A special case of contrast stretching
where ? ? 0 is called clipping. This is
useful for noise reduction when the input signal
is known to lie in the range a, b. Clipping
also should be performed on images which will be
represented with a finite number of bits, for
example, with unsigned character.
4
Clipping function
Thresholding function
- Thresholding Thresholding is a special case of
clipping where a b t and the output becomes
binary. - Image Negative
5
- examples of contrast stretching
6
- examples of clipping and thresholding
7
- examples of image negative
8
- Range Compression The dynamic range of a
typical unitarily transformed image is so large
that only a few pixels are visible. The dynamic
range can be compressed via the logarithmic
transformation.
- Image subtraction In many imaging applications,
it is desired to compare two images. A simple but
powerful method is to align the two images and
subtract them.
where (i,j) means each pixel position, and u1 and
u2 are two images being compared.
9
- Histogram Modification The histogram of a
digital image represents the relative frequency
of occurrence of each graylevel in the image. The
histogram gives an estimate of the probability of
occurrence of each graylevel. Histogram
equalization In histogram equalization, the goal
is to obtain a uniform histogram for the output
image. Consider image pixel value u?0 to be a
random variable with a continuous probability
density function pu(u). Then the random variable
CDF
will be uniformly distributed over (0,1). From
(1), the derivative of v with respect to u is
10
Applying the relation between pv(v) and pu(u) by
the transformation
That is, the transformed pdf, pv(v), has a
uniform pdf. - Implementing
11
Ex. 1) Histogram equalization for the given
histogram h(u) of a 3-bit image.
Histogram Specification The goal of histogram
specification is mapping a distribution into a
given specific histogram. Suppose the random
variable u?0 with probability density pu(u) to be
transformed to v?0 such that it has a specified
probability density pv(v). For this to be true,
we define a uniform random variable
12
that also satisfies the relation
Eliminating w, we obtain
- Implementing
13
Ex. 2) Histogram specification for the given
histograms(pdf) pu(u) and pv(v) for a 2-bit image.
14
- image characteristics according to histogram
patterns
15
- examples of histogram equalization
16
- example of histogram specification
histogram equalization
original image
histogram specification
histograms