20

尝试将大小为 100x100 的灰度图像分割成大小为 39x39 的重叠块,步幅大小为 1。这意味着下一个从右侧/或下方开始一个像素的块仅与上一个块不同一个额外的列/或行。

代码粗略:首先计算每个补丁的索引,以便能够从图像构建补丁的二维数组并能够从补丁构建图像:

patches = imgFlat[ind]

“补丁”是一个二维数组,每列包含一个向量形式的补丁。

这些补丁被处理,每个补丁单独并随后再次合并到图像中,并使用预先计算的索引。

img = np.sum(patchesWithColFlat[ind],axis=2)

由于补丁重叠,最后有必要将 img 与预先计算的权重相乘:

imgOut = weights*imgOut

我的代码真的很慢,速度是一个关键问题,因为这应该在 ca 上完成。10^8 个补丁。

函数 get_indices_for_un_patchify 和 weights_unpatchify 可以预先计算一次,因此速度只是 patchify 和 unpatchify 的问题。

感谢任何提示。

卡洛斯

import numpy as np
import scipy
import collections
import random as rand


def get_indices_for_un_patchify(sImg,sP,step):
    ''' creates indices for fast patchifying and unpatchifying

    INPUTS:
      sx    image size
      sp    patch size
      step  offset between two patches (default == [1,1])

      OUTPUTS:
       patchInd             collection with indices
       patchInd.img2patch   patchifying indices
                            patch = img(patchInd.img2patch);
       patchInd.patch2img   unpatchifying indices

    NOTE: * for unpatchifying necessary to add a 0 column to the patch matrix
          * matrices are constructed row by row, as normally there are less rows than columns in the 
            patchMtx
     '''
    lImg = np.prod(sImg)
    indImg = np.reshape(range(lImg), sImg)

    # no. of patches which fit into the image
    sB = (sImg - sP + step) / step

    lb              = np.prod(sB)
    lp              = np.prod(sP)
    indImg2Patch    = np.zeros([lp, lb])
    indPatch        = np.reshape(range(lp*lb), [lp, lb])

    indPatch2Img = np.ones([sImg[0],sImg[1],lp])*(lp*lb+1)

    # default value should be last column
    iRow   = 0;
    for jCol in range(sP[1]):
        for jRow in range(sP[0]):
            tmp1 = np.array(range(0, sImg[0]-sP[0]+1, step[0]))
            tmp2 = np.array(range(0, sImg[1]-sP[1]+1, step[1]))
            sel1                    = jRow  + tmp1
            sel2                    = jCol  + tmp2
            tmpIndImg2Patch = indImg[sel1,:]          
            # do not know how to combine following 2 lines in python
            tmpIndImg2Patch = tmpIndImg2Patch[:,sel2]
            indImg2Patch[iRow, :]   = tmpIndImg2Patch.flatten()

            # next line not nice, but do not know how to implement it better
            indPatch2Img[min(sel1):max(sel1)+1, min(sel2):max(sel2)+1, iRow] = np.reshape(indPatch[iRow, :, np.newaxis], sB)
            iRow                    += 1

    pInd = collections.namedtuple
    pInd.patch2img = indPatch2Img
    pInd.img2patch = indImg2Patch

    return pInd

def weights_unpatchify(sImg,pInd):
    weights = 1./unpatchify(patchify(np.ones(sImg), pInd), pInd)
    return weights

# @profile
def patchify(img,pInd):
    imgFlat = img.flat
   # imgFlat = img.flatten()
    ind = pInd.img2patch.tolist()
    patches = imgFlat[ind]

    return patches

# @profile
def unpatchify(patches,pInd):
    # add a row of zeros to the patches matrix    
    h,w = patches.shape
    patchesWithCol = np.zeros([h+1,w])
    patchesWithCol[:-1,:] = patches
    patchesWithColFlat = patchesWithCol.flat
   #  patchesWithColFlat = patchesWithCol.flatten()
    ind = pInd.patch2img.tolist()
    img = np.sum(patchesWithColFlat[ind],axis=2)
    return img

我在这里调用这些函数,例如使用随机图像

if __name__ =='__main__':
    img = np.random.randint(255,size=[100,100])
    sImg = img.shape
    sP = np.array([39,39])  # size of patch
    step = np.array([1,1])  # sliding window step size
    pInd = get_indices_for_un_patchify(sImg,sP,step)
    patches = patchify(img,pInd)
    imgOut = unpatchify(patches,pInd)
    weights = weights_unpatchify(sImg,pInd)
    imgOut = weights*imgOut

    print 'Difference of img and imgOut = %.7f' %sum(img.flatten() - imgOut.flatten())
4

1 回答 1

23

“修补”数组的一种有效方法,即获取原始数组的窗口数组是创建一个具有自定义 strides 的视图即跳转到下一个元素的字节数。将 numpy 数组视为(美化的)内存块可能会有所帮助,然后 strides 是将索引映射到内存地址的一种方法。

例如,在

a = np.arange(10).reshape(2, 5)

a.itemsize等于 4(即,每个元素 4 个字节或 32 位)并且a.strides(20, 4)(5 个元素,1 个元素),因此a[1,2]指的是在第一个元素之后是1*20 + 2*4字节(或元素)的元素:1*5 + 2

0 1 2 3 4
5 6 7 x x

实际上,元素是一个接一个地放入内存中,0 1 2 3 4 5 6 7 x x但是跨步让我们将其索引为二维数组。

基于这个概念,我们可以重写patchify如下

def patchify(img, patch_shape):
    img = np.ascontiguousarray(img)  # won't make a copy if not needed
    X, Y = img.shape
    x, y = patch_shape
    shape = ((X-x+1), (Y-y+1), x, y) # number of patches, patch_shape
    # The right strides can be thought by:
    # 1) Thinking of `img` as a chunk of memory in C order
    # 2) Asking how many items through that chunk of memory are needed when indices
    #    i,j,k,l are incremented by one
    strides = img.itemsize*np.array([Y, 1, Y, 1])
    return np.lib.stride_tricks.as_strided(img, shape=shape, strides=strides)

这个函数返回一个视图img,所以没有分配内存,它只运行了几十微秒。输出形状并不完全是您想要的,实际上必须复制它才能获得该形状。

在处理比基本数组大得多的数组视图时必须小心,因为操作会触发需要分配大量内存的副本。在您的情况下,由于阵列不是太大并且没有那么多补丁,所以应该没问题。

最后,我们可以稍微整理一下补丁数组:

patches = patchify(img, (39,39))
contiguous_patches = np.ascontiguousarray(patches)
contiguous_patches.shape = (-1, 39**2)

这不会重现您的 patchify 函数的输出,因为您是按 Fortran 顺序开发补丁的。我建议您改用它,因为

  1. 它会在以后导致更自然的索引(即,对于您的解决方案,第一个补丁是补丁 [0] 而不是补丁 [:, 0])。

  2. 在 numpy 中使用 C 排序也更容易,因为你需要更少的输入(你避免像 order='F' 这样的东西,默认情况下,数组是按 C 顺序创建的......)。

如果您坚持使用“提示”:strides = img.itemsize * np.array([1, Y, Y, 1]),请使用.reshape(..., order='F')contiguous_patches最终转置它.T

于 2013-05-28T09:39:56.750 回答