Convert point cloud data to COPC format¶

Convert .laz to .copc.laz using pdal library, following the workflow in this blog post, although in the blog post they convert from the native projection to web mercator before converting to COPC. It's not necessary to reproject the data. So we just use one pipeline to convert the original .laz to COPC.

In [1]:

import pdal
import fsspec

In [2]:

fs_read = fsspec.filesystem('s3', anon=True)

In [3]:

url = 's3://cmgp-sfm-public-read-bucket/RSCC_EventResponse_Jan2023/20230105_MB_SC_Cap_UltraHi_NAD83_2011_UTM10N_ConfWater_Trimmed.laz'
fs_read.info(url)

Out[3]:

{'ETag': '"9787cc8cdd4fcea93c73de5ce2e89b80-332"',
 'LastModified': datetime.datetime(2023, 1, 9, 22, 14, 5, tzinfo=tzutc()),
 'size': 2784154775,
 'name': 'cmgp-sfm-public-read-bucket/RSCC_EventResponse_Jan2023/20230105_MB_SC_Cap_UltraHi_NAD83_2011_UTM10N_ConfWater_Trimmed.laz',
 'type': 'file',
 'StorageClass': 'STANDARD',
 'VersionId': None,
 'ContentType': 'binary/octet-stream'}

In [4]:

#fs_read.ls('s3://cmgp-sfm-public-read-bucket/RSCC_EventResponse_Jan2023/LAZ_for_Release_2023_1_11/')  # bucket doesn't allow anon listing 

In [5]:

from pathlib import Path
p = Path(url)
local_laz_file = f'{p.stem}.laz'
local_laz_file

Out[5]:

'20230105_MB_SC_Cap_UltraHi_NAD83_2011_UTM10N_ConfWater_Trimmed.laz'

In [6]:

fs_local = fsspec.filesystem('file')

In [7]:

if not fs_local.exists(local_laz_file):
    fs_read.download(url, local_laz_file)

Convert LAZ to COPC¶

In [8]:

local_copc_file = f"{p.stem}.6339.copc.laz"

write_copc_pipeline = [
    {
        "type":"readers.las",
        "filename":local_laz_file
    },
    {
        "type":"writers.copc",
        "filename":local_copc_file
    }
]

In [9]:

write_copc_pipeline = str(write_copc_pipeline).replace("'", '"')

In [10]:

pipeline = pdal.Pipeline(write_copc_pipeline)
count = pipeline.execute()

Upload to ESIP public read bucket¶

In [11]:

fs_write = fsspec.filesystem('s3', profile='esip-qhub')

In [12]:

s3_copc_file = f's3://esip-qhub-public/testing/{local_copc_file}'

_ = fs_write.upload(local_copc_file, s3_copc_file)

Check the uploaded file¶

In [13]:

fs_read.info(s3_copc_file)

Out[13]:

{'ETag': '"1018c2cfec1dff7d3ba5ebc3e2f5bbb3-56"',
 'LastModified': datetime.datetime(2023, 1, 12, 16, 14, 43, tzinfo=tzutc()),
 'size': 2914701303,
 'name': 'esip-qhub-public/testing/20230105_MB_SC_Cap_UltraHi_NAD83_2011_UTM10N_ConfWater_Trimmed.6339.copc.laz',
 'type': 'file',
 'StorageClass': 'STANDARD',
 'VersionId': None,
 'ContentType': 'binary/octet-stream'}

View with COPC viewer (Click the viewer_output URL produced by this cell):¶

In [14]:

viewer_url = f'https://viewer.copc.io/?copc=https://esip-qhub-public.s3.amazonaws.com/testing/{local_copc_file}'
print(viewer_url)

https://viewer.copc.io/?copc=https://esip-qhub-public.s3.amazonaws.com/testing/20230105_MB_SC_Cap_UltraHi_NAD83_2011_UTM10N_ConfWater_Trimmed.6339.copc.laz

Notes on run environment¶

This notebook was run on the ESIP Nebari deployment (JupyterHub on Kubernetes) at https://nebari.esipfed.org, using the 32GB server type (it exceeded memory with the 8GB or 16GB server types). We used this conda environment:

name: pdal
channels:
- conda-forge
dependencies:
- python=3.10
- pdal
- python-pdal
- fsspec
- s3fs
- ipykernel

In [ ]: