More Point Operations

 

Home
Image Fundamentals
Fourier Transforms
Fourier Properties
Point Operations
More Point Operations
Spatial Filters
Frequency Filters
Image Restoration
Frequency Filters
Homomorphic Filters
Color Models
Color Palettes
Color Processing
Image Geometry
Image Compression
Run Length Encoding
Lossy Compression

Jan 31, 2001

 

Histogram Specification

Equalization produces a more-or-less uniform histogram. Sometimes it is desired to modify the input to meet a specified output histogram.

1.  Equalize the input image:

2.  Given the desired output histogram, get the CDF that equalizes the histogram:  

                               

3.  Invert this last equation: 

                       

 4.  Finally, replace v with s.

Contrast Stretching

Contrast stretching (also known as normalization or autoscaling) is a method of enhancing image contrast which is more controllable than histogram equalization. The latter is a fully specified operation with no user modifications, while the former allows user specification of the input and output intensity ranges.

Let [a,b] be the lower and upper limits, respectively, of the output intensity levels, and [c,d] the corresponding limits of the input image. Then the desired mapping function is

Selective contrast control
Logarithmic LUT enhances low intensities and reduces spread of high intensities
Exponential LUT enhances high intensities and reduces spread of low intensities

Image Arithmetic
Addition: Images can be summed to reduce random noise
Subtraction: Images can be subtracted to remove uneven background and to show variation in pictures.
Multiplication: Images can be multiplied to correct for uneven sensor response

 

Last modified on January 31, 2001