把图像划分为patch以及用图像块重建图像

github上有了现成的代码： 

GitHub - dovahcrow/patchify.py: A library that helps you split image into small, overlappable patches, and merge patches into original image.A library that helps you split image into small, overlappable patches, and merge patches into original image. - GitHub - dovahcrow/patchify.py: A library that helps you split image into small, overlappable patches, and merge patches into original image.https://github.com/dovahcrow/patchify.py该package可以直接通过pip安装。

安装：

pip install patchify

使用：

（1）把图像划分为图像块（patch）

patchify(image_to_patch, patch_shape, step=1)

示例1：2D image

#This will split the image into small images of shape [3,3]
patches = patchify(image, (3, 3), step=1)

示例2：3D image:

#This will split the image into small images of shape [3,3,3]
patches = patchify(image, (3, 3, 3), step=1)

（2）把patch拼接回原始图像

unpatchify(patches_to_merge, merged_image_size)

示例：

reconstructed_image = unpatchify(patches, image.shape)

This will reconstruct the original image that was patchified in previous code.

完整的示例

2D 图像分块与拼接

import numpy as np
from patchify import patchify, unpatchify

image = np.array([[1,2,3,4], [5,6,7,8], [9,10,11,12]])

patches = patchify(image, (2,2), step=1) # split image into 2*3 small 2*2 patches.

assert patches.shape == (2, 3, 2, 2)
reconstructed_image = unpatchify(patches, image.shape)

assert (reconstructed_image == image).all()

3D 图像分块与拼接

import numpy as np
from patchify import patchify, unpatchify

image = np.random.rand(512,512,3)

patches = patchify(image, (2,2,3), step=1) # patch shape [2,2,3]
print(patches.shape) # (511, 511, 1, 2, 2, 3). Total patches created: 511x511x1

assert patches.shape == (511, 511, 1, 2, 2, 3)
reconstructed_image = unpatchify(patches, image.shape)
print(reconstructed_image.shape) # (512, 512, 3)

assert (reconstructed_image == image).all()