Matplotlib网格计数图导出KML后显示异常，求可行解决方案

我已成功生成可正常显示的Matplotlib网格计数图，但将其导出为KML格式后，使用Google Earth打开时网格计数无法正确显示（相关截图已附上）。请问是否可以通过KML格式正确展示网格计数？

实现代码 #

import geopandas as gpd
import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
import shapely
import contextily as cx
import simplekml
from operator import itemgetter

df = pd.read_csv('https://raw.githubusercontent.com/plotly/datasets/master/Nuclear%20Waste%20Sites%20on%20American%20Campuses.csv')

gdf = gpd.GeoDataFrame(df, geometry=gpd.points_from_xy(df.lon, df.lat))
gdf.crs = "EPSG:4326"

fig = plt.figure(1)
ax = fig.add_axes([0,0,1,1])
ax.axis('off')

plt.axis('off')
border1 = plt.axis()

#if False:
#    plt.show()
#else:
#    pngName = 'Overlay.png'
#    fig.savefig(pngName, facecolor=fig.get_facecolor(), transparent=False)

# total area for the grid
#xmin, ymin, xmax, ymax = gdf.total_bounds

bounds = gdf.geometry.apply(lambda x: x.bounds).tolist()
xmin, ymin, xmax, ymax = min(bounds, key=itemgetter(0))[0], min(bounds, key=itemgetter(1))[1], max(bounds, key=itemgetter(2))[2], max(bounds, key=itemgetter(3))[3]

# how many cells across and down
n_cells = 10
cell_size = (xmax-xmin)/n_cells

# projection of the grid
#crs = "+proj=sinu +lon_0=0 +x_0=0 +y_0=0 +a=6371007.181 +b=6371007.181 +units=m +no_defs"

# create the cells in a loop
grid_cells = []
for x0 in np.arange(xmin, xmax+cell_size, cell_size ):
    for y0 in np.arange(ymin, ymax+cell_size, cell_size):
        # bounds
        x1 = x0-cell_size
        y1 = y0+cell_size
        grid_cells.append(shapely.geometry.box(x0, y0, x1, y1))
        
cell = gpd.GeoDataFrame(grid_cells, columns = ['geometry'], crs = 'EPSG:4326')

merged = gpd.sjoin(gdf, cell, how='left', predicate='within')

# make a simple count variable that we can sum
merged['gridCount'] = 1

# Compute stats per grid cell -- aggregate fires to grid cells with dissolve
dissolve = merged.dissolve(by = 'index_right', aggfunc = 'count')

# put this into cell
cell.loc[dissolve.index, 'gridCount'] = dissolve['gridCount'].values

ax = cell.plot(column = 'gridCount', 
               figsize = (12, 8), 
               cmap = 'viridis_r', 
               vmax = 1000,
               alpha = 0.5,
               edgecolor = 'grey')
   
cx.add_basemap(ax, zoom = 8, crs = 'EPSG:4326')

pngName = 'Overlay.png'
plt.axis('off')
plt.savefig(pngName, facecolor=fig.get_facecolor(), transparent=False)

bottomleft  = xmin,ymin
bottomright = xmax,ymin
topright    = xmax,ymax
topleft     = xmin,ymax

kml = simplekml.Kml()
ground = kml.newgroundoverlay(name='GroundOverlay')
ground.icon.href = pngName
ground.gxlatlonquad.coords =[bottomleft, bottomright, topright, topleft]
kml.save("GroundOverlay.kml")

可以通过KML正确展示网格计数，当前图片叠加方案失效的核心原因是投影匹配误差和Google Earth对栅格叠加的渲染限制，推荐两种可靠实现方式：

方式1：直接导出矢量网格到KML #

放弃图片叠加，将带计数属性的网格矢量数据直接导出，Google Earth可原生渲染矢量并支持点击查看计数，无变形问题：

# 方法A：用geopandas直接导出（需libkml依赖）
cell.to_file("grid_count.kml", driver="KML")

# 方法B：用simplekml手动构建带样式的网格（兼容无libkml环境）
kml = simplekml.Kml()
cmap = plt.get_cmap('viridis_r')

for idx, row in cell.iterrows():
    count = row.get('gridCount', 0)
    # 归一化计数并转换为KML支持的AARRGGBB颜色格式
    norm_count = min(count / 1000, 1)
    rgba = cmap(norm_count)
    kml_color = simplekml.Color.rgb(int(rgba[0]*255), int(rgba[1]*255), int(rgba[2]*255), int(rgba[3]*255))
    
    # 创建多边形网格
    poly = kml.newpolygon(name=f"网格{idx} | 计数:{count}")
    poly.outerboundaryis.coords = list(row.geometry.exterior.coords)
    # 设置填充色和边框
    poly.style.polystyle.color = kml_color
    poly.style.polystyle.outline = 1

kml.save("grid_count_vector.kml")

方式2：修复图片叠加的投影匹配问题 #

若坚持用图片叠加，需解决两个关键问题：

1. 清除图片空白边：原代码保存图片时会生成额外空白，导致和地理边界错位：

# 替换原图片保存代码
plt.gca().set_aspect('equal', adjustable='box')
plt.savefig(pngName, bbox_inches='tight', pad_inches=0, transparent=True)

2. 修正KML坐标顺序：Google Earth的gx:LatLonQuad要求坐标按左下角→右下角→右上角→左上角→左下角闭合，确保无拉伸：

# 修复坐标顺序并添加闭合点
ground.gxlatlonquad.coords = [bottomleft, bottomright, topright, topleft, bottomleft]
ground.gxlatlonquad.rotation = 0  # 强制无旋转

注意事项 #

• 矢量方案比图片叠加更可靠，支持交互且无投影变形；
• 图片叠加必须严格保证图片宽高比与地理区域经纬度跨度比一致，否则会出现拉伸；
• 全程使用EPSG:4326（WGS84）投影，Google Earth原生支持该投影，避免使用其他投影导致匹配误差。

内容的提问来源于stack exchange，提问作者Chopin