在数字图像处理中，filters.rank 子模块（属于 scikit-image 库）
提供了一系列基于局部邻域排序或形态学运算的非线性滤波方法。
与传统线性滤波（如高斯滤波、均值滤波）不同，
这些方法需要用户自定义设定滤波器的形状和大小，
因此需要导入 morphology 模块来设定。
autolevel
autolevel 这个词在 Photoshop 里面翻译成自动色阶，用局部直方图来对图片进行滤波分级。
该滤波器局部地拉伸灰度像素值的直方图，以覆盖整个像素值范围。
格式：
skimage.filters.rank.autolevel(image,selem)selem 表示结构化元素，用于设定滤波器。
In [2]:
%matplotlib inline
from skimage import data,color
import matplotlib.pyplot as plt
from skimage.morphology import disk
import skimage.filters.rank as sfr
img =color.rgb2gray(data.coffee())
半径为5的圆形滤波器：
In [3]:
auto =sfr.autolevel(img, disk(5))
/opt/conda/lib/python3.12/site-packages/IPython/core/interactiveshell.py:3579: UserWarning: Possible precision loss converting image of type float64 to uint8 as required by rank filters. Convert manually using skimage.util.img_as_ubyte to silence this warning. exec(code_obj, self.user_global_ns, self.user_ns)
In [4]:
plt.figure('filters',figsize=(8,8))
plt.subplot(121)
plt.title('origin image')
plt.imshow(img,plt.cm.gray)
Out[4]:
<matplotlib.image.AxesImage at 0x7f62d664b6e0>
In [5]:
plt.subplot(122)
plt.title('filted image')
plt.imshow(auto,plt.cm.gray)
Out[5]:
<matplotlib.image.AxesImage at 0x7f62cc077980>
In [6]:
from skimage import data,color
import matplotlib.pyplot as plt
from skimage.morphology import disk
from skimage.morphology import black_tophat, white_tophat
img =color.rgb2gray(data.coffee())
半径为5的圆形滤波器：
In [7]:
auto = black_tophat(img, disk(5))
In [8]:
plt.figure('filters',figsize=(8,8))
plt.subplot(121)
plt.title('origin image')
plt.imshow(img,plt.cm.gray)
Out[8]:
<matplotlib.image.AxesImage at 0x7f6315466150>
In [9]:
plt.subplot(122)
plt.title('filted image')
plt.imshow(auto,plt.cm.gray)
Out[9]:
<matplotlib.image.AxesImage at 0x7f62cbf023c0>
In [10]:
from skimage import data,color
import matplotlib.pyplot as plt
from skimage.morphology import disk
import skimage.filters.rank as sfr
img =color.rgb2gray(data.coffee())
半径为5的圆形滤波器：
In [11]:
auto =sfr.enhance_contrast(img, disk(5))
/opt/conda/lib/python3.12/site-packages/IPython/core/interactiveshell.py:3579: UserWarning: Possible precision loss converting image of type float64 to uint8 as required by rank filters. Convert manually using skimage.util.img_as_ubyte to silence this warning. exec(code_obj, self.user_global_ns, self.user_ns)
In [12]:
plt.figure('filters',figsize=(8,8))
plt.subplot(121)
plt.title('origin image')
plt.imshow(img,plt.cm.gray)
Out[12]:
<matplotlib.image.AxesImage at 0x7f62cbf78a10>
In [13]:
plt.subplot(122)
plt.title('filted image')
plt.imshow(auto,plt.cm.gray)
Out[13]:
<matplotlib.image.AxesImage at 0x7f62cbf1a1b0>
In [14]:
from skimage import data,color
import matplotlib.pyplot as plt
from skimage.morphology import disk
import skimage.filters.rank as sfr
img =color.rgb2gray(data.coffee())
半径为5的圆形滤波器：
In [15]:
dst =sfr.entropy(img, disk(5))
/opt/conda/lib/python3.12/site-packages/IPython/core/interactiveshell.py:3579: UserWarning: Possible precision loss converting image of type float64 to uint8 as required by rank filters. Convert manually using skimage.util.img_as_ubyte to silence this warning. exec(code_obj, self.user_global_ns, self.user_ns)
In [16]:
plt.figure('filters',figsize=(8,8))
plt.subplot(121)
plt.title('origin image')
plt.imshow(img,plt.cm.gray)
Out[16]:
<matplotlib.image.AxesImage at 0x7f62cc09db50>
In [17]:
plt.subplot(122)
plt.title('filted image')
plt.imshow(dst,plt.cm.gray)
Out[17]:
<matplotlib.image.AxesImage at 0x7f62cbef2090>
In [18]:
from skimage import data,color
import matplotlib.pyplot as plt
from skimage.morphology import disk
import skimage.filters.rank as sfr
img =color.rgb2gray(data.coffee())
半径为5的圆形滤波器：
In [19]:
dst =sfr.equalize(img, disk(5))
/opt/conda/lib/python3.12/site-packages/IPython/core/interactiveshell.py:3579: UserWarning: Possible precision loss converting image of type float64 to uint8 as required by rank filters. Convert manually using skimage.util.img_as_ubyte to silence this warning. exec(code_obj, self.user_global_ns, self.user_ns)
In [20]:
plt.figure('filters',figsize=(8,8))
plt.subplot(121)
plt.title('origin image')
plt.imshow(img,plt.cm.gray)
Out[20]:
<matplotlib.image.AxesImage at 0x7f62cbd48fb0>
In [21]:
plt.subplot(122)
plt.title('filted image')
plt.imshow(dst,plt.cm.gray)
Out[21]:
<matplotlib.image.AxesImage at 0x7f62cbdae0f0>
In [22]:
from skimage import data,color
import matplotlib.pyplot as plt
from skimage.morphology import disk
import skimage.filters.rank as sfr
img =color.rgb2gray(data.coffee())
半径为5的圆形滤波器：
In [23]:
dst =sfr.gradient(img, disk(5))
/opt/conda/lib/python3.12/site-packages/IPython/core/interactiveshell.py:3579: UserWarning: Possible precision loss converting image of type float64 to uint8 as required by rank filters. Convert manually using skimage.util.img_as_ubyte to silence this warning. exec(code_obj, self.user_global_ns, self.user_ns)
In [24]:
plt.figure('filters',figsize=(8,8))
plt.subplot(121)
plt.title('origin image')
plt.imshow(img,plt.cm.gray)
Out[24]:
<matplotlib.image.AxesImage at 0x7f62cbf4fd40>
In [25]:
plt.subplot(122)
plt.title('filted image')
plt.imshow(dst,plt.cm.gray)
Out[25]:
<matplotlib.image.AxesImage at 0x7f62cbc36150>
dst =sfr.maximum(img, disk(5)) 最小值滤波器（minimum) ：返回图像局部区域内的最小值，用此最小值取代该区域内所有像素值。
dst =sfr.minimum(img, disk(5))均值滤波器（mean): 返回图像局部区域内的均值，用此均值取代该区域内所有像素值。
dst =sfr.mean(img, disk(5)) 中值滤波器（median): 返回图像局部区域内的中值，用此中值取代该区域内所有像素值。
dst =sfr.median(img, disk(5))莫代尔滤波器（modal) : 返回图像局部区域内的modal值，用此值取代该区域内所有像素值。
dst =sfr.modal(img, disk(5))otsu阈值滤波（otsu): 返回图像局部区域内的otsu阈值，用此值取代该区域内所有像素值。
dst =sfr.otsu(img, disk(5))阈值滤波（threshhold): 将图像局部区域中的每个像素值与均值比较，大于则赋值为1，小于赋值为0，得到一个二值图像。
dst =sfr.threshold(img, disk(5)) 减均值滤波（subtract_mean): 将局部区域中的每一个像素，减去该区域中的均值。
dst =sfr.subtract_mean(img, disk(5))求和滤波（sum):求局部区域的像素总和，用此值取代该区域内所有像素值。
dst =sfr.sum(img, disk(5))