the shine toolbox matlab tools for controlling low-level image properties

Post on 01-Jan-2016

298 Views

Category:

Documents

0 Downloads

Preview:

Click to see full reader

TRANSCRIPT

The SHINE toolbox

Matlab tools for controlling low-level image properties

SHINE functionsSHINE (Spectrum, Histogram, and Intensity Normalization and Equalization)

lumMatch• Function to equate images in terms of mean luminance and

contrast (i.e., the standard deviation of the luminance distribution)• newimages = lumMatch(images,mask,lum)• face1new = (face1-mean2(face1))/std2(face1)*S+M;

mean2(face1)= 181.8557

mean2(face2)= 203.6699

std2(face1)= 85.8567

std2(face2)= 61.5167

mean2(face1new)= 192.4541

mean2(face2new)= 192.7071

std2(face1)= 73.4324

std2(face2)= 73.3302

INPUT OUTPUT

histMatch

• Function that matches the luminance histograms across images

• newimages = histMatch(images,optim,hist,mask)

• See ‘help histMatch’ for infos about the input parameters

sfMatch

• Function that equates the energy at each spatial frequency (i.e., the rotational average of the Fourier amplitude spectra)

• newimages = sfMatch(images,rescaling,tarmag)

specMatch

• Function for equating the Fourier amplitude spectra across images

• images = specMatch(images,rescaling,tarmag)

Simultaneous matching

• The matching of histograms and Fourier amplitudes affect each other in a way that histograms e.g. are not equated anymore after specMatch

• Joint equalization of low-level properties using an iterative approach

Download

www.mapageweb.umontreal.ca/gosselif/shine

top related