Home       Trees       Indices       Help   
Package pyvision :: Package other :: Module normalize
[hide private]
[frames] | no frames]

Module normalize

source code

Functions [hide private]
 
normalizeMeanStd(matrix)
TODO: deprecated please use meanStd.		 source code
 
clipRange(matrix, min_val, max_val)
zero mean, one standard deviation		 source code
 
meanStd(matrix)
zero mean, one standard deviation		 source code
 
meanUnit(matrix)
zero mean, unit length		 source code
 
unit(matrix)
unit length		 source code
 
selfQuotientImage(matrix, sigma=5.0)
Compute a self quotient image.		 source code
 
gaussianFilter(im, sigma)
Smooth an image using a Gaussian filter.		 source code
 
highPassFilter(matrix, sigma)
This function computes a high and low pass filter.		 source code
 
lowPassFilter(matrix, sigma)
This function computes a low pass filter.		 source code
 
bandPassFilter(matrix, sigma_low, sigma_high)
This function computes a high and low pass filter.		 source code
Variables [hide private]
  __package__ = 'pyvision.other'
Function Details [hide private]

selfQuotientImage(matrix, sigma=5.0)

source code 

Compute a self quotient image.

Based on work by Wang et.al. "Self Quotient Image for Face Recognition" ICIP 2004

highPassFilter(matrix, sigma)

source code 

This function computes a high and low pass filter. This can be used to reduce the effect of lighting.

A low pass image is first computed by convolving the image with a Gausian filter of radius sigma. Second, a high pass image is computed by subtracting the low pass image from the original image. This means that the original image can be reconstructed by adding a low pass image and a high pass image.

Returns:
high_pass_image

lowPassFilter(matrix, sigma)

source code 

This function computes a low pass filter. It basically smoothes the image by convolving with a Gaussian. This is often used to reduce the effect of noise in images or to reduce the effect of small registration errors.

Returns:
an pv.Image set from a numpy matrix if input was an image or a numpy matrix otherwize.

bandPassFilter(matrix, sigma_low, sigma_high)

source code 

This function computes a high and low pass filter. This can be used to reduce the effect of lighting.

A low pass image is first computed by convolving the image with a Gausian filter of radius sigma. Second, a high pass image is computed by subtracting the low pass image from the original image. This means that the original image can be reconstructed by adding a low pass image and a high pass image.

Returns:
high_pass_image

   Home       Trees       Indices       Help   
Generated by Epydoc 3.0.1 on Tue Apr 21 14:31:58 2015 http://epydoc.sourceforge.net