如何将一系列图像绘制到特定地图上

Question

我想以特定模式绘制图像，如下图所示

我想了解使用 python 绘制图像的最佳方法是什么。我使用以下方法以网格模式绘制图像。 输出看起来像这样

import numpy as np
import matplotlib.pyplot as plt
from datetime import datetime
import matplotlib as mpl
from matplotlib.dates import date2num
import matplotlib.dates as mdates
from mpl_toolkits.axes_grid1 import ImageGrid

img0 = f['B00'][i_ant, :, :]
img1 = f['B01'][i_ant, :, :]
img2 = f['B02'][i_ant, :, :]
img3 = f['B03'][i_ant, :, :]
    

img_arr = [img0,img1,img2,img3]
fig = plt.figure(figsize=(5., 5.))
grid = ImageGrid(fig, 111, 
                     nrows_ncols=(2, 2),  # creates 2x2 grid of axes
                     axes_pad=0,  # pad between axes
                     )

for ax, im in zip(grid, img_arr):
    ax.imshow(im)
        

plt.show()

Answer 1

对于方形螺旋的每一层，都会添加两条边，除了最后一层只需要一条边。 i,j 位置每一步递增或递减 1，并且在每个拐角处方向旋转 90 度。

这就是它的工作方式：

import matplotlib.pyplot as plt
import numpy as np

N = 6
fig = plt.figure(figsize=(16, 18), constrained_layout=True)
spec = fig.add_gridspec(ncols=N * 2 - 1, nrows=N * 2 - 1)
i, j = N - 2 + N % 2, N - 1
dir_i, dir_j = 1, - 1
plot_num = 0
for k in range(1, N + 1):
    for _ in range(2):  # add two strokes of k subplots (only 1 when k==N)
        for _ in range(k):
            ax = fig.add_subplot(spec[i, j])
            ax.imshow(np.random.rand(3, 3))
            ax.set_xlabel(f'{plot_num} [{k}]', fontsize=18)
            plot_num += 1
            i += dir_i
            j += dir_j
        if k == N:
            break
        dir_i, dir_j = -dir_j, dir_i
plt.show()

要准确复制原始编号，可以绘制一些同心正方形。对于奇数N，最后一个方格很不规则，只有两条边，最后一行跳到对面，比其他方格少一个方格。

对于奇怪的 2 和 3 编号，可以引入一些重新编号：

import matplotlib.pyplot as plt
import numpy as np

N = 6
fig = plt.figure(figsize=(16, 18), constrained_layout=True)
spec = fig.add_gridspec(ncols=2 * N - 1, nrows=2 * N - 1)
plot_num = 0
for k in range(2, N + 2, 2):
    # i, j = N - 2 + N % 2, N - 1
    i, j = N - k + N % 2, N - 1
    dir_i, dir_j = 1, - 1
    for side in range(4):  # add four strokes of k subplots (only 2 when k==N)
        for _ in range(k - 1):
            ax = fig.add_subplot(spec[i, j])
            modified_plot_num = 5 - plot_num if plot_num in (2, 3) else plot_num
            ax.imshow(np.random.rand(6, 6), cmap='inferno')
            ax.set_xlabel(f'{modified_plot_num} [{k}]', fontsize=18)
            plot_num += 1
            i += dir_i
            j += dir_j
            if plot_num == N * N:  # for odd N, the very last cell should be skipped
                break
        if k == N + 1:
            if side == 0:  # for last side of uneven square: jump to the other side
                dir_i, dir_j = -dir_i, -dir_j
                i, j = i - 1, 2 * N - 2
            elif side == 1:
                break
        dir_i, dir_j = -dir_j, dir_i
plt.show()