Use LAZ as reference data

[naniciafone]

We need to ensure that reference LAZ can be provided. Currently, IRC uses Ames Stereo Pipeline to make a LAZ out of a reference GeoTIFF, where the center of every pixel is converted to a point. The tool then uses Iterative Closest Point to align source and GeoTIFF-derived reference point clouds. See (https://github.com/cryogars/ice-road-copters/blob/main/scripts/laz_align.py) line 75, and literature about pc_align here. We want to maintain the capability of using a reference GeoTIFF incase that is all the user has. Use branch issue-23-LAZ. It's possible reference point cloud can be provided, we need to test.

Tasks

• Research (maybe test) if COPC, EPT, or LAZ is more efficiently processed by IRC [EDIT: Per Dom's advice, COPC is LAZ.
• If reference point cloud is provided, we will want to run that point cloud through the same filters as the source data OR create a flag to not do this. This is because filtering/ segmentation algorithm should be consistent across REF/SOURCE data.

[Ibrahim-Ola]

Noted!

[naniciafone]

I am transforming MCS reference (bare earth) LAZ to EPSG 32611 (it was provided by QSI in EPSG 6340+5703), which is what we export our collected data to. Once I do this, I'll upload to RDE Base and we can use these LAZ and a snow-on flight as source and reference LAZ for testing.

[naniciafone]

@granitehills14 @cwagnar

RDE base paths:
Path for reference LAZ: \LIDAR_Airborne\2021-MCS-QSI\MCS_WGS84
And snow-on: \LIDAR_Airborne\IDALS

[Ibrahim-Ola]

@naniciafone has approved that I post a step-by-step guide to making contributions on GitHub, in case anyone needs it.

Step-by-Step Git Contribution Guide

Note: All commands in this guide should be executed on your terminal/command prompt.

Setup

1. Install Git (if you don't already have it)

Git comes pre-installed on most operating systems. However, if you do not have Git, you can download and install it from https://git-scm.com/downloads. If you are unsure whether you have Git or not, open your terminal/command prompt and run:

git --version

2. Configure Your Git Identity

This configuration helps you track your activities (e.g., commits).

git config user.name "Your Name"
git config user.email "your-email@example.com"

Important: Use the same email that's verified in your GitHub account! Username can be anything, but I strongly recommend using your GitHub username so teammates can easily identify your contributions.

3. Install Conda (if you don't already have it)

If you don’t have Conda installed, download Anaconda or MiniConda. See Installing Conda to know which installer is right for you. Other options are available for non-conda users. For example, you can create a virtual environment with venv and manage Python versions with pyenv.

4. Clone the Repository

Go to the folder where you want to house ice-road-copters (IRC) and clone the repository locally using the command below:

git clone git@github.com:cryogars/ice-road-copters.git
cd ice-road-copters # changes directory to IRC

5. Install the Conda Environment

The creators of IRC provided a .yaml file, which is a Conda environment that enables us to install the dependent Python packages with a one-liner.

conda env create -f iceroad_env.yaml
conda activate iceroad # this activates the IRC Conda environment

Working on an Existing Branch

For now, we will all be collaborating on the same branch. Follow the steps below to contribute:

6. See All Available Branches

git fetch origin  # Download latest branch info
git branch -a     # Then see all available branches

7. Switch to the Branch You Want to Work On

git checkout branch-name

Example: git checkout issue-23-LAZ

8. Pull Latest Changes (in case others worked on it)

git pull origin branch-name

Example: git pull origin issue-23-LAZ

Making Changes

9. Make Your Code Changes

• Edit files using your preferred editor.
• Save your changes.

10. Check What You've Changed

git status

This shows which files you've modified.

11. Add Your Changes

git add .

This stages all your changes. Or add specific files: git add filename.py

12. Commit Your Changes

git commit -m "Brief description of what you changed"

Example: git commit -m "Fix lidar processing bug in LAZ reader"

13. Push Your Changes

git push origin branch-name

Example: git push origin issue-23-LAZ

Important Notes

• Always work on a branch - never work directly on main
• Pull before you push - Step 8 prevents conflicts.
• Write clear commit messages - Describe what you actually changed.
• Ask for help if you get stuck (spoiler, we all do!).

Common Commands Summary

# Get updates
git fetch origin
git pull origin branch-name

# Check status/branches
git status
git branch -a

# Make changes
git add .
git commit -m "Your message"
git push origin branch-name

[naniciafone]

Fantastic, thanks @Ibrahim-Ola !!

[naniciafone]

@Ibrahim-Ola If the user does NOT want to run the reference point cloud through the PDAL filtering pipeline, we need to make sure that only ground points (class 2) from reference LAZ are used for co-registration and differencing. As discussed with Josh, we want the default to be that the reference point cloud IS run through the filtering pipeline (

ice-road-copters/scripts/laz2dem.py

Line 61 in ac5cade

def create_json_pipeline(in_fp, outlas, outtif, dem_fp, json_name = 'las2unaligned', json_dir = './jsons', canopy = False):

). This ensures that the ground segmentation process is the same across reference and source data. Also, the DEM filter will need to be turned off if the reference data is LAZ.

[naniciafone]

Also, output raster resolution comes from the reference raster resolution, so we will need the user input the desired resolution. They will need to know what quality 3dep product they have, or the point density of their input LAZ. We can make a note about this in the readme.

[jomey]

@Ibrahim-Ola
Could we put your guide into a repo CONTRIBUTING.md guide?
Think this is a great resources that should be kept for future reference outside this issue

[Ibrahim-Ola]

@jomey Will do!