GLCM图像中的黑色空间 [英] Black space in GLCM image

问题描述

我正在尝试使用Haralick描述的GLCM(能量，均匀性等)对我拥有的一系列4波段(R，G，B，NIR)航拍图像进行一些质地测量.我已经在一个子集上尝试过了，但是最终得到的图像大部分是空白的.我目前的理解是，它与灰度和levels参数有关，但我无法弄清楚.

I'm trying to calculate some textural measures using the GLCM described by Haralick (energy, homogeneity, etc.) for a series of 4 band (R, G, B, NIR) aerial photographs that I have. I have tried this on a subset but am ending up with an image that is mostly blank. My current understanding is that it has to do with the greyscaling and the levels parameter but I can't figure it out.

我的日期非常大(几个GB)，所以我试图通过使用模块RIOS来提高效率(以400倍400倍nbands numpy数组的形式读取图像，处理数据并将其写出到输出图像中).

My date is very large (several GB) so I'm trying to be efficient by using the module RIOS (reads an image in as a 400×400×nbands numpy array, processes the data and writes out to an output image).

可以在此处找到我的输入场景(200 MB).

My input scene can be found here (200 MB).

我的输出图像看起来像(由于黑色像素非常小，因此可能很难看到):

My output image looks like (this may be difficult to see as the black pixels are very small):

我的代码是:

#Set up input and output filenames
infiles = applier.FilenameAssociations()
infiles.image1 = "infile.tif"

outfiles = applier.FilenameAssociations()
outfiles.outimage = "outfile.tif"

controls = applier.ApplierControls()
controls.progress = cuiprogress.CUIProgressBar()
# I ultimately want to use a window here
# which RIOS easily allows you to set up.
# For a 3x3 the overlap is 1, 5x5 overlap is 2 etc
#controls.setOverlap(4)

def doFilter(info, infiles, outfiles, controls=controls):
    grayImg = img_as_ubyte(color.rgb2gray(infiles.image1[3]))
    g = greycomatrix(grayImg, distances=[1], angles=[0, np.pi/4, np.pi/2, 3*np.pi/4], symmetric=True, normed=True)
    filtered = greycoprops(g, 'energy')
    # create 3d image from 2d array
    outfiles.outimage = numpy.expand_dims(filtered, axis=0)


applier.apply(doFilter, infiles, outfiles, controls=controls)

很显然这里出了点问题，因为我的输出不符合我的预期.我的猜测是与"levels"参数有关.我已经在这里指出了一个解释: GLCM结果中的黑线可以很好地解释参数，但是我无法改善结果.

Obviously there is something wrong here as my output is not as I expect. My guess that it is to do with the 'levels' parameter. I have been pointed to an explanation here: Black line in GLCM result which explains the parameter well but I am unable to improve my result.

有人可以向我解释为什么我的结果显示出来了，如何补救?

Can someone explain to me why my result is coming out as shown and how I can remedy it?

推荐答案

下面的代码计算与tif图像NIR波段的偏移量向上1像素偏移量向上"相对应的GLCM:

The code below computes the GLCM corresponding to an offset "1-pixel offset upwards" from the NIR band of your tif image:

import numpy as np
from skimage import io
from skimage.feature import greycomatrix, greycoprops

x = io.imread('m_2909112_se_15_1_20150826.tif')
nir = x[:, :, 3]

glcm = greycomatrix(nir, [1], [np.pi/2], levels=256, normed=True, symmetric=True)

nir的外观如下:

将参数normed设置为True的效果是将计算出的GLCM除以其总和，结果，glcm的元素具有相当小的值.这是一个示例:

The effect of setting the parameter normed to True is that the computed GLCM is divided by its total sum, and as a result the elements of glcm have rather small values. Here's a sample:

In [48]: np.set_printoptions(precision=3)

In [49]: glcm[:5, :5, 0, 0]
Out[49]: 
array([[  0.000e+00,   0.000e+00,   0.000e+00,   0.000e+00,   0.000e+00],
       [  0.000e+00,   2.725e-03,   6.940e-05,   3.725e-05,   2.426e-05],
       [  0.000e+00,   6.940e-05,   1.709e-04,   4.103e-05,   2.216e-05],
       [  0.000e+00,   3.725e-05,   4.103e-05,   4.311e-04,   4.222e-05],
       [  0.000e+00,   2.426e-05,   2.216e-05,   4.222e-05,   5.972e-05]])

要将glcm显示为图像，您需要对其重新缩放，例如:

To display glcm as an image you need to rescale it, for example like this:

from skimage.exposure import rescale_intensity
scaled = rescale_intensity(glcm[:,:,0,0])
io.imshow(scaled)

这篇关于GLCM图像中的黑色空间的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！