Last modified: October 21, 2013
Contents
FloatVector fdbroken_a ()
Operates on: | Image [OneBit] |
---|---|
Returns: | FloatVector |
Category: | Features |
Defined in: | broken.py |
Author: | Christian Brandt and Christoph Dalitz |
Fourier descriptor for broken shapes according to Eq.(19) in [Dalitz2013]. The coefficient c_0 is used for scale normalisation.
The absolute values of the coefficients A(0), A(N-1) A(1), A(N-2), ... are returned.
Reference:
[Dalitz2013] (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17) Dalitz, C., Brandt, C., Goebbels, S., Kolanus, D.: "Fourier Descriptors for Broken Shapes." EURASIP Journal on Advances in Signal Processing 2013:161, 2013
Note
In this study, fdbroken_a was the best performing Fourier descriptor on the NEUMES data set of broken shapes.
FloatVector fdbroken_a_phase ()
Operates on: | Image [OneBit] |
---|---|
Returns: | FloatVector |
Category: | Features |
Defined in: | broken.py |
Author: | Christian Brandt and Christoph Dalitz |
Fourier descriptor for broken shapes according to Eq. (19) in [Dalitz2013]. The coefficients c_r and c_s are used for scale and phase normalisation. c_r and c_s are the two coefficients with the largest and second largest value (r >= 0, 0 < s < N/2).
The complex values of the coefficients A(0), A(N/2-1) A(1), A(N/2-2), ... are returned as alternating real and imaginary parts.
FloatVector fdbroken_a_phase_s1 ()
Operates on: | Image [OneBit] |
---|---|
Returns: | FloatVector |
Category: | Features |
Defined in: | broken.py |
Author: | Christian Brandt and Christoph Dalitz |
Fourier descriptor for broken shapes according to Eq. (19) in [Dalitz2013]. The coefficients c_0 and c_1 are used for scale and phase normalisation.
The complex values of the coefficients A(0), A(N/2-1) A(1), A(N/2-2), ... are returned as alternating real and imaginary parts.
FloatVector fdbroken_b ()
Operates on: | Image [OneBit] |
---|---|
Returns: | FloatVector |
Category: | Features |
Defined in: | broken.py |
Author: | Christian Brandt and Christoph Dalitz |
Fourier descriptor for broken shapes according to Eq. (20) in [Dalitz2013].
The values of the coefficients B(1), B(2), ... B(N) are returned.
FloatVector fdbroken_c ()
Operates on: | Image [OneBit] |
---|---|
Returns: | FloatVector |
Category: | Features |
Defined in: | broken.py |
Author: | Christian Brandt and Christoph Dalitz |
Fourier descriptor for broken shapes according to Eq. (22) in [Dalitz2013]. The coefficient c_0 is used for scale normalisation.
The absolute values of the coefficients C(0), ... C(N/2-1) are returned.
FloatVector fdbroken_c_phase ()
Operates on: | Image [OneBit] |
---|---|
Returns: | FloatVector |
Category: | Features |
Defined in: | broken.py |
Author: | Christian Brandt and Christoph Dalitz |
Fourier descriptor for broken shapes according to Eq. (22) in [Dalitz2013]. The coefficients c_r and c_s are used for scale and phase normalisation. c_r and c_s are the two coefficients with the largest and second largest absolute value (r >= 0, 0 < s < N/2).
The complex values of the coefficients C(0), ... C(N/4-1) are returned as alternating real and imaginary parts.
FloatVector fdbroken_c_phase_s1 ()
Operates on: | Image [OneBit] |
---|---|
Returns: | FloatVector |
Category: | Features |
Defined in: | broken.py |
Author: | Christian Brandt and Christoph Dalitz |
Fourier descriptor for broken shapes according to Eq. (22) in [Dalitz2013]. The coefficients c_0 = 1/N sum(r(t)) is used for scale normalisation. The coefficient c_1 is used for phase normalization.
The complex values of the coefficients C(0), ... C(N/4-1) are returned as alternating real and imaginary parts.
FloatVector fdsingle_centroid_distance ()
Operates on: | Image [OneBit] |
---|---|
Returns: | FloatVector |
Category: | Features |
Defined in: | single.py |
Author: | Christian Brandt and Christoph Dalitz |
Centroid distance Fourier descriptor for unbroken shapes according to Eq. (16) in [Dalitz2013]. The coefficient c_0 is used for scale normalisation.
The absolute values of the coefficients R(1), R(2), ... are returned.
FloatVector fdsingle_centroid_distance_phase ()
Operates on: | Image [OneBit] |
---|---|
Returns: | FloatVector |
Category: | Features |
Defined in: | single.py |
Author: | Christian Brandt and Christoph Dalitz |
Centroid distance Fourier descriptor for unbroken shapes according to Eq. (16) in [Dalitz2013]. The coefficients c_0 and c_s are used for scale and phase normalisation. c_s is the coefficient with the largest absolute value (s >= 1).
The complex values of the coefficients R(1), R(2), ... are returned as alternating real and imaginary parts.
FloatVector fdsingle_complex_position ()
Operates on: | Image [OneBit] |
---|---|
Returns: | FloatVector |
Category: | Features |
Defined in: | single.py |
Author: | Christian Brandt and Christoph Dalitz |
Complex position Fourier descriptor for unbroken shapes according to Eq. (6) in [Dalitz2013]. The coefficient c_r is used for scale normalisation. c_r is the coefficient with the largest absolute value.
The absolute values of the coefficients l(1), l(N-1), l(2), l(N-2), ... are returned.
Note
In the study [Dalitz2013], this was the best performing Fourier descriptor on the MPEG7 data set of connected shapes.
FloatVector fdsingle_complex_position_phase ()
Operates on: | Image [OneBit] |
---|---|
Returns: | FloatVector |
Category: | Features |
Defined in: | single.py |
Author: | Christian Brandt and Christoph Dalitz |
Complex position Fourier descriptor for unbroken shapes according to Eq. (6) in [Dalitz2013]. The coefficients c_r and c_s are used for scale and phase normalisation. c_r and c_s are the two coefficients with the largest and second largest absolute value. (r >= 0, 0 < s < N/2).
The complex values of the coefficients l(1), l(N-1) l(2), l(N-2), ... are returned as alternating real and imaginary parts.
FloatVector fdsingle_complex_position_phase_r1 ()
Operates on: | Image [OneBit] |
---|---|
Returns: | FloatVector |
Category: | Features |
Defined in: | single.py |
Author: | Christian Brandt and Christoph Dalitz |
Complex position Fourier descriptor for unbroken shapes according to Eq. (6) in [Dalitz2013]. The coefficients c_1 and c_r are used for scale and phase normalisation. c_r is the coefficient with the largest absolute value.
The complex values of the coefficients l(1), l(N-1) l(2), l(N-2), ... are returned as alternating real and imaginary parts.
FloatVector fdsingle_complex_position_r1 ()
Operates on: | Image [OneBit] |
---|---|
Returns: | FloatVector |
Category: | Features |
Defined in: | single.py |
Author: | Christian Brandt and Christoph Dalitz |
Complex position Fourier descriptor for unbroken shapes according to Eq. (6) in [Dalitz2013]. The coefficient c_1 is used for scale normalisation.
The absolute values of the coefficients l(1), l(N-1), l(2), l(N-2), ... are returned.
FloatVector fdsingle_elliptic ()
Operates on: | Image [OneBit] |
---|---|
Returns: | FloatVector |
Category: | Features |
Defined in: | single.py |
Author: | Christian Brandt and Christoph Dalitz |
Elliptic Fourier descriptor for unbroken shapes according to Eq. (12) in [Dalitz2013]. The coefficient I_1 is used for normalisation.
The values of the coefficients I(1), J(1), k1(1), I(2), J(2), K(2), ... are returned.
FloatVector fdsingle_granlund ()
Operates on: | Image [OneBit] |
---|---|
Returns: | FloatVector |
Category: | Features |
Defined in: | single.py |
Author: | Christian Brandt and Christoph Dalitz |
Granlund's Fourier descriptor for unbroken shapes according to Eq. (7) in [Dalitz2013]. The coefficient c_1 is used for normalisation. c_r is the coefficient with the largest absolute value.
The complex values of the coefficients d(2), d(3), ... are returned as alternating real and imaginary parts.
FloatVector fdsingle_real_position ()
Operates on: | Image [OneBit] |
---|---|
Returns: | FloatVector |
Category: | Features |
Defined in: | single.py |
Author: | Christian Brandt and Christoph Dalitz |
Real position Fourier descriptor for unbroken shapes according to Eq. (13) in [Dalitz2013]. The coefficient c_1 is used for normalisation.
The values of the coefficients with indexes starting at 1 are returned.