python scipy卷積運(yùn)算的實(shí)現(xiàn)方法

>>> import numpy as np
>>> x=np.array([1,2,3])
>>> h=np.array([4,5,6])
>>> import scipy.signal
>>> scipy.signal.convolve(x,h) #卷積運(yùn)算
array([ 4, 13, 28, 27, 18])

>>> import numpy as np
>>> from scipy import signal
>>> import matplotlib.pyplot as plt
>>> sig=np.random.randn(1000) #生成隨機(jī)數(shù)
>>> autocorr=signal.fftconvolve(sig,sig[::-1],mode='full') #fft算法實(shí)現(xiàn)卷積
>>> fig,(ax_orig,ax_mag)=plt.subplots(2,1) #建立兩行一列圖形
>>> ax_orig.plot(sig) #在第一行把原始的隨機(jī)數(shù)序列sig畫(huà)出來(lái)
[<matplotlib.lines.Line2D object at 0x0000000006E1DC88>]
>>> ax_orig.set_title('White noise') #設(shè)置標(biāo)題'白噪聲'
<matplotlib.text.Text object at 0x0000000006931860>
>>> ax_mag.plot(np.arange(-len(sig)+1,len(sig)),autocorr) #卷積后的圖像
[<matplotlib.lines.Line2D object at 0x0000000006E1DB00>]
>>> ax_mag.set_title('Autocorrelation') #設(shè)置標(biāo)題
<matplotlib.text.Text object at 0x0000000006DFE8D0>
>>> fig.tight_layout() #此句可以防止圖像重疊
>>> fig.show() #顯示圖像

>>> import numpy as np
>>> from scipy import signal
>>> from scipy import misc
>>> import matplotlib.pyplot as plt
>>> face=misc.face(gray=True) #創(chuàng)建一個(gè)灰度圖像
>>> scharr=np.array([[-3-3j,0-10j,+3-3j],
    [-10+0j,0+0j,+10+0j],
     [-3+3j,0+10j,+3+3j]]) #設(shè)置一個(gè)特殊的卷積和
>>> grad=signal.convolve2d(face,scharr,boundary='symm',mode='same') #把圖像的face數(shù)組和設(shè)計(jì)好的卷積和作二維卷積運(yùn)算,設(shè)計(jì)邊界處理方式為symm
>>> fig,(ax1,ax2)=plt.subplots(1,2,figsize=(10,6)) #建立1行2列的圖fig
>>> ax1.imshow(face,cmap='gray') #顯示原始的圖
<matplotlib.image.AxesImage object at 0x00000000078FC198>
>>> ax1.set_axis_off() #不顯示坐標(biāo)軸
>>> ax2.imshow(np.absolute(grad),cmap='gray') #顯示卷積后的圖
<matplotlib.image.AxesImage object at 0x00000000078FCE48>
>>> ax2.set_axis_off() #不顯示坐標(biāo)軸
>>> fig.show() #顯示繪制好的畫(huà)布