matlab PHASECONG3函数解析

大佬网站的链接：PHASECONG3

这个函数是用来计算图像的边缘和角的相位一致性

下面是该函数的注释
This function calculates the PC_2 measure of phase congruency.
PC_2就是相位一致性
This function supersedes PHASECONG2 and PHASECONG being faster and requires
less memory
第三版，比前两版更快，内存更少
There are potentially many arguments, here is the full usage:
有很多潜在参数，下面是完整用法
[M m or ft pc EO, T] = phasecong3(im, nscale, norient, minWaveLength, mult, sigmaOnf, k, cutOff, g, noiseMethod)

However, apart from the image, all parameters have defaults and the usage can be as simple as:
除了image，其他都不是必须的
M = phasecong3(im);

Arguments Default values Description

nscale           4    - Number of wavelet scales, try values 3-6，小波数量，默认4
norient          6    - Number of filter orientations. 过滤器方向的数量
minWaveLength    3    - Wavelength of smallest scale filter.最小尺度的滤波器波长
mult             2.1  - Scaling factor between successive filters.连续滤波器之间的比例因子
sigmaOnf         0.55 - Ratio of the standard deviation of the Gaussian 
                        describing the log Gabor filter's transfer function 
                        in the frequency domain to the filter center frequency.
						描述对数Gabor滤波器传递函数的高斯函数在频域上的标准差与滤波器中心频率的比值。
k                2.0  - No of standard deviations of the noise energy beyond
                        the mean at which we set the noise threshold point.
                        You may want to vary this up to a value of 10 or
                        20 for noisy images
						噪声能量超过我们设定的噪声阈值点的平均值的标准偏差。
cutOff           0.5  - The fractional measure of frequency spread
                        below which phase congruency values get penalized.
						频率扩展的分数度量，低于它的相位一致性值会受到惩罚
g                10   - Controls the sharpness of the transition in
                        the sigmoid function used to weight phase
                        congruency for frequency spread.  
						控制sigmoid函数中转换的锐度，用于加权频率扩展的相位一致性。
noiseMethod      -1   - Parameter specifies method used to determine
                        noise statistics. 
                          -1 use median of smallest scale filter responses
                          -2 use mode of smallest scale filter responses
                           0+ use noiseMethod value as the fixed noise threshold 
						参数指定用于确定噪声统计的方法。 

Returned values:
M - Maximum moment of phase congruency covariance.
This is used as a indicator of edge strength.
相位一致性协方差的最大矩。这被用作边缘强度的指标。
m - Minimum moment of phase congruency covariance.
This is used as a indicator of corner strength.
相位一致性协方差的最小矩。这被用作拐角强度的指标。
or - Orientation image in integer degrees 0-180,
positive anticlockwise.
0 corresponds to a vertical edge, 90 is horizontal.
图像方位范围在0-180度，逆时针为正方向。0对应垂直边，90对应水平边。
ft - Local weighted mean phase angle at every point in the
image. A value of pi/2 corresponds to a bright line, 0
corresponds to a step and -pi/2 is a dark line.
图像中每一点的局部加权平均相位角。/2的值对应一条亮线，0对应一个阶跃，- /2是一条暗线
pc - Cell array of phase congruency images (values between 0 and 1)
for each orientation
每个方向的相位一致性图像(值在0和1之间)的单元阵列
EO - A 2D cell array of complex valued convolution result
一个复值卷积结果的二维单元数组
T - Calculated noise threshold (can be useful for
diagnosing noise characteristics of images). Once you know
this you can then specify fixed thresholds and save some
computation time.
计算噪声阈值(可用于诊断图像的噪声特征)。一旦您知道了这一点，您就可以指定固定的阈值并节省一些计算时间。

EO{s,o} = convolution result for scale s and orientation o. The real part
is the result of convolving with the even symmetric filter, the imaginary
part is the result from convolution with the odd symmetric filter.
卷积结果是{刻度s,方向o}，实部是与偶对称滤波器卷积的结果，虚部是与奇对称滤波器卷积的结果

Hence:
abs(EO{s,o}) returns the magnitude of the convolution over the
image at scale s and orientation o.
angle(EO{s,o}) returns the phase angles.
因此abs(EO{s,o})返回在尺度s和方向o的图像上卷积的大小。angle(EO{s,o})返回相位角。

Notes on specifying parameters:

The parameters can be specified as a full list eg.

[M m or ft pc EO] = phasecong3(im, 5, 6, 3, 2.5, 0.55, 2.0, 0.4, 10);

or as a partial list with unspecified parameters taking on default values

[M m or ft pc EO] = phasecong3(im, 5, 6, 3);

or as a partial list of parameters followed by some parameters specified via a
keyword-value pair, remaining parameters are set to defaults, for example:

[M m or ft pc EO] = phasecong3(im, 5, 6, 3, ‘cutOff’, 0.3, ‘k’, 2.5);

The convolutions are done via the FFT. Many of the parameters relate to the
specification of the filters in the frequency plane. The values do not seem
to be very critical and the defaults are usually fine. You may want to
experiment with the values of ‘nscales’ and ‘k’, the noise compensation factor.
卷积是通过FFT完成的。许多参数与频率平面滤波器的规格有关。这些值似乎不是很重要，缺省值通常都很好。你可以用“nscales”和“k”的值做实验，这是噪声补偿因子。

Notes on filter settings to obtain even coverage of the spectrum
sigmaOnf .85 mult 1.3
sigmaOnf .75 mult 1.6 (filter bandwidth ~1 octave)
sigmaOnf .65 mult 2.1
sigmaOnf .55 mult 3 (filter bandwidth ~2 octaves)

See Also: PHASECONG, PHASECONG2, PHASESYM, GABORCONVOLVE, PLOTGABORFILTERS