如何在不杀死内核的情况下将多个TIF文件与Rioxarray Python合并

发布于 2025-01-26 06:13:45 字数 1661 浏览 2 评论 0原文

我正在使用RGB飓风IDA的高RES NOAA航空影像数据。数据可在下载下可用 - 任何日期时间都将下载大量TIFS https://storms.ngs.noaa.gov/storms/ida/index.html#18/29.46140/-90.30946

我想沿着（x，y，y，y，y，y，y，y，y，y，y，y，频段）为了创建一个完整的“马赛克”，以识别任何缺失的图像贴片，然后创建一个构建轮廓面具。

我有一个可以合并20个TIF的功能，但最重要的是杀死了内核。有更好的方法可以进一步合并吗？理想情况下，图像中想要40。

import rioxarray
from rioxarray import merge
from rasterio.plot import show 
 
def combine_tif_large(file_list, title=''):
        """Plot the combined image of a set of tif files - chunks long file list into groups of 50 (n) to prevent datasets closing
        file_list = list of file paths for files to be merged
        title = title for plot 
        """
        n = 50
        fig, ax = plt.subplots( figsize=(20,20))
        chunked_files = [file_list[i:i + n] for i in range(0, len(file_list), n)]
    
        final = [] 
        for chunk in chunked_files:
            now = datetime.now()
            current_time = now.strftime("%H:%M:%S")
            print("Current Time =", current_time)
            print('starting chunk..')
            elements = []
            for val in chunk:
                elements.append(rioxarray.open_rasterio(val))
            print('finish chunk')
            print('starting  merge')
            merged = merge.merge_arrays(elements, nodata=0.0)
            final.append(merged)
        print('starting final merge')
        merge_final = merge.merge_arrays(final, nodata=0.0)
        image = merge_final.values
        show(image, ax = ax, title= title)

原文

I am working with high res NOAA aerial imagery data from Hurrican Ida, RGB. Data is available here under download - any of the date times will download a large amount of tifs https://storms.ngs.noaa.gov/storms/ida/index.html#18/29.46140/-90.30946

I want to merge multiple tifs along (x , y , bands) in order to create a complete 'mosaic', to identify any missing patches of imagery, and then create a building outline mask.

I have a function that can merge 20 Tifs, but above that it kills the kernel. Is there a better way to merge further? Ideally would want 40 in an image.

import rioxarray
from rioxarray import merge
from rasterio.plot import show 
 
def combine_tif_large(file_list, title=''):
        """Plot the combined image of a set of tif files - chunks long file list into groups of 50 (n) to prevent datasets closing
        file_list = list of file paths for files to be merged
        title = title for plot 
        """
        n = 50
        fig, ax = plt.subplots( figsize=(20,20))
        chunked_files = [file_list[i:i + n] for i in range(0, len(file_list), n)]
    
        final = [] 
        for chunk in chunked_files:
            now = datetime.now()
            current_time = now.strftime("%H:%M:%S")
            print("Current Time =", current_time)
            print('starting chunk..')
            elements = []
            for val in chunk:
                elements.append(rioxarray.open_rasterio(val))
            print('finish chunk')
            print('starting  merge')
            merged = merge.merge_arrays(elements, nodata=0.0)
            final.append(merged)
        print('starting final merge')
        merge_final = merge.merge_arrays(final, nodata=0.0)
        image = merge_final.values
        show(image, ax = ax, title= title)

分享到QQ

分享到微博