Add noise to XLD contours.
add_noise_white_contour_xld adds noise to the input XLD contours Contours and returns the noisy contours in NoisyContours. For each point along the original contours the local regression line (i.e., a least-squares approximating line) based on NumRegrPoints neighboring points is determined. Then the point is shifted perpendicular to this line. The length of the shifts corresponds to white noise, equally distributed in the interval [-Amp,Amp] generated by using the C function ``drand48'' with an initial time dependent seed.
|
Contours (input_object) |
xld_cont(-array) -> object |
| Original contours. | |
|
NoisyContours (output_object) |
xld_cont(-array) -> object |
| Noisy contours. | |
|
NumRegrPoints (input_control) |
integer -> integer |
| Number of points used to calculate the regression line. | |
| Default value: 5 | |
| Suggested values: 3, 5, 7, 9 | |
| Restriction: (NumRegrPoints >= 3) && odd(NumRegrPoints) | |
|
Amp (input_control) |
real -> real |
| Maximum amplitude of the added noise (equally distributed in [-Amp,Amp]). | |
| Default value: 1.0 | |
| Suggested values: 0.25, 0.5, 1.0, 1.5, 2.0, 3.0, 4.0, 5.0, 10.0 | |
| Restriction: Amp > 0 | |
draw_ellipse(WindowHandle,Row,Column,Phi,Radius1,Radius2)
gen_ellipse_contour_xld(Ellipse,Row,Column,Phi,Radius1,Radius2,0,6.28319,
'positive')
add_noise_white_contour_xld(Ellipse,NoisyEllipse,5,0.5)
fit_ellipse_contour_xld (NoisyEllipse, 'fitzgibbon', -1, 2, 0, 200, 3, 2.0,
ERow, EColumn, EPhi, ERadius1, ERadius2, EStartPhi,
EEndPhi, 'EPointOrder).
add_noise_white_contour_xld returns 2 (H_MSG_TRUE) if all parameter values are correct. If the input is empty the behaviour can be set via set_system('no_object_result',<Result>). If necessary, an exception is raised.
add_noise_white_contour_xld is reentrant, local, and processed without parallelization.
gen_contours_skeleton_xld, lines_gauss, lines_facet, edges_sub_pix, gen_ellipse_contour_xld
smooth_contours_xld, add_noise_white
Sub-pixel operators
