def __init__(self,block_size = 7, aperture_size=3, k=0.04, **kwargs):
pass
DetectorROI.__init__(self,**kwargs)
self.block_size = block_size
self.aperture_size = aperture_size
self.k = k
def __init__(self,block_size = 7, aperture_size=3, k=0.04, **kwargs):
pass
DetectorROI.__init__(self,**kwargs)
self.block_size = block_size
self.aperture_size = aperture_size
self.k = k
def __init__(self, sigma=DEFAULT_SIGMA, scales=DOG_SCALES, min_size=20, min_contrast=0.03, max_curvature_ratio=10, **kwargs):
'''
min_size - Image pyramid terminates when one of the image demensions reaches this size.
'''
DetectorROI.__init__(self,**kwargs)
self.min_size = min_size
self.scales = scales
self.sigma = sigma
self.min_contrast = min_contrast
self.max_curvature_ratio = max_curvature_ratio
def __init__(self, sigma=DEFAULT_SIGMA, scales=DOG_SCALES, min_size=20, min_contrast=0.03, max_curvature_ratio=10, **kwargs):
'''
min_size - Image pyramid terminates when one of the image demensions reaches this size.
'''
DetectorROI.__init__(self,**kwargs)
self.min_size = min_size
self.scales = scales
self.sigma = sigma
self.min_contrast = min_contrast
self.max_curvature_ratio = max_curvature_ratio
def __init__(self,
mask=[[-1, 0, 1]],
radius=9,
sigma=0.7,
k=0.04,
**kwargs):
'''
Corner Detector
mask - first dirivitive filter
radius - radius of the max filter
sigma - sigma of the smoothing gaussian.
k - not sure what this parameter means.
Passed to superclass:
n - is the approximate number of points requested.
bin_size - the width and height of each bin in pixels.
corner_selector ('all', 'bins', or 'best') - stratagy for point selection.
When corner_selector is set to bins, the image is subdivided in to bins of
size <bin_size>X<bin_size> pixels and an equal number of points will be taken
from each of those bins. This insures that points are found in all parts of the
image not just where the corners are strongest.
This code is based on a function originally written for matlab.
Original matlab code by:
Jingyu Yan and Marc Pollefeys
Department of Computer Science
University of North Carolina at Chapel Hill
Converted to Python by:
David Bolme
Department of Computer Science
Colorado State Univerisity
'''
DetectorROI.__init__(self, **kwargs)
self.mask = mask
self.radius = radius
self.sigma = sigma
self.k = k
def __init__(self,filter = [[-1,0,1]], radius=9, sigma=0.7, k=0.04, **kwargs):
'''
Corner Detector
filter - first dirivitive filter
radius - radius of the max filter
sigma - sigma of the smoothing gaussian.
k - not sure what this parameter means.
Passed to superclass:
n - is the approximate number of points requested.
bin_size - the width and height of each bin in pixels.
corner_selector ('all', 'bins', or 'best') - stratagy for point selection.
When corner_selector is set to bins, the image is subdivided in to bins of
size <bin_size>X<bin_size> pixels and an equal number of points will be taken
from each of those bins. This insures that points are found in all parts of the
image not just where the corners are strongest.
This code is based on a function originally written for matlab.
Original matlab code by:
Jingyu Yan and Marc Pollefeys
Department of Computer Science
University of North Carolina at Chapel Hill
Converted to Python by:
David Bolme
Department of Computer Science
Colorado State Univerisity
'''
DetectorROI.__init__(self,**kwargs)
self.filter = filter
self.radius = radius
self.sigma = sigma
self.k = k
def __init__(self, min_hessian=400.0, **kwargs):
'''
'''
self.min_hessian = min_hessian
DetectorROI.__init__(self, **kwargs)
def __init__(self, min_hessian=400.0, **kwargs):
"""
"""
self.min_hessian = min_hessian
DetectorROI.__init__(self, **kwargs)