components.photo.vignettecorrector
|
Vignette corrector. |
Vignette analysis. |
- class VignetteCorrector(config={}, **kwds)[source]
Bases:
Transformer
Vignette corrector.
Adjust the brightness of images with a radially varying gain function. This should be applied to ‘linear intensity’ image data before gamma correction is applied.
The
mode
parameter sets the function to use. Theparam_n
values set how the correction varies with radius. Their meaning depends on the function. TheAnalyseVignette
component can be used to generate optimised values.Config
mode
str
Function to use. Possible values:
'power'
,'poly2'
,'poly3'
,'lin2'
,'lin3'
,'invlin2'
,'invlin3'
.param_0
float
First function parameter.
param_1
float
Second function parameter.
param_2
float
Third function parameter.
param_3
float
Fourth function parameter.
param_4
float
Fifth function parameter.
- class AnalyseVignette(config={}, **kwds)[source]
Bases:
Transformer
Vignette analysis.
Measures the average luminance of 50 circular bands of an input grey image, then calculates the optimum function parameters to correct it.
The
mode
configuration selects the function to fit. If set tomeasure
orinv_measure
no function is fitted. Ininv_measure
mode the output shows the vignetting instead of the required correction. Available functions are:- power
1.0 + (a * (r ^ b))
- poly2
1.0 + (a * (r ^ 2)) + (b * (r ^ 4))
- poly3
1.0 + (a * (r ^ 2)) + (b * (r ^ 4)) + (c * (r ^ 6))
- lin2
2-segment piecewise linear
- lin3
3-segment piecewise linear
- invlin2
inverse 2-segment piecewise linear
- invlin3
inverse 3-segment piecewise linear
The
function
output emits the measured and fitted gain values. It can be connected to aPlotData
component.The
plot...
parameters can be used to control the plot’s appearance. Running the component network several times with different options allows quite complex plots to be built up.Config
mode
str
Function to fit. Possible values:
'measure'
,'inv_measure'
,'power'
,'poly2'
,'poly3'
,'lin2'
,'lin3'
,'invlin2'
,'invlin3'
.method
str
Curve fitting method:
lm
,trf
, ordogbox
.plot_measurement
bool
Include the measured input in the plot.
plot_error
bool
Include the residual error in the plot.
plot_label_measured
str
Label for the ‘measured’ plot.
plot_label_fitted
str
Label for the ‘fitted’ plot. If left blank
mode
is used.- outputs = ['output', 'function']
Comments or questions? Please email jim@jim-easterbrook.me.uk.