https://github.com/precision-sustainable-ag/SemiF-Preprocessing/tree/feature/reimplementation

Pre-Requisites

NCSU Unity ID and password
Duo 2-Factor Authentication Enabled
1. https://ncsu.service-now.com/sp?id=kb_article&sys_id=175315d7db8573802fa0791c8c961926
Access to SUNNY
An internet connection
Cisco Secure Global Connect (NCSU’s VPN Client)
1. https://ncsu.service-now.com/sp?id=kb_article_view&sys_kb_id=54b5a9f9470062182360891e436d4369
2. https://ncsu.service-now.com/sp/en?id=kb_article&sys_id=73c7e352db5844d084129ac5db9619fe

Setting Up Your Machine

SSH Setup

PowerShell

make sure that PowerShell is installed
- PowerShell is automatically installed on Windows 10 and higher - it’s also available on MacOS and Linux
- If you don’t have Windows PowerShell, install it using winget in Command Prompt
  - winget install Microsoft.PowerShell
https://learn.microsoft.com/en-us/windows-server/administration/openssh/openssh_install_firstuse?tabs=powershell&pivots=windows-server-2025#enable-openssh-for-windows-server-2025
https://www.ionos.com/digitalguide/server/configuration/powershell-ssh/

Open Windows PowerShell
To enable SSH capabilities you will need administrator privileges. To run PowerShell as an administrator run the command Start-Process powershell -Verb RunAs in PowerShell
In the window that opens run Start-Service sshd
1. If an error occurs saying that the service does not exist run the command Add-WindowsCapability -Online -Name OpenSSH.Server~~~~0.0.1.0
2. Then run Start-Service sshd
We recommend running Set-Service -Name sshd -StartupType 'Automatic' to automatically start SSHD to make sure it stays enabled in the future

Finally, run the following command to verify that the SSHD setup process automatically configured the firewall rule:

if (!(Get-NetFirewallRule -Name "OpenSSH-Server-In-TCP" -ErrorAction SilentlyContinue | Select-Object Name, Enabled)) {
    Write-Output "Firewall Rule 'OpenSSH-Server-In-TCP' does not exist, creating it..."
    New-NetFirewallRule -Name 'OpenSSH-Server-In-TCP' -DisplayName 'OpenSSH Server (sshd)' -Enabled True -Direction Inbound -Protocol TCP -Action Allow -LocalPort 22
} else {
    Write-Output "Firewall rule 'OpenSSH-Server-In-TCP' has been created and exists."
}

You can now use ssh to setup remote connection to SUNNY in PowerShell

Visual Studio Code

Install Visual Studio Code https://code.visualstudio.com/
Install Remote-SSH Extension https://marketplace.visualstudio.com/items?itemName=ms-vscode-remote.remote-ssh

Setup Remote Access to SUNNY

Login to NCSU’s VPN

Open Cisco Secure Global Connect on your computer
Type vpn.ncsu.edu into the blank field
Click Connect
Login to the VPN using your NCSU credentials
1. login: <unityID>
2. password: <password>
3. second password: <method of 2-Factor Authentication>
  1. The default method is push and leaving the field blank or typing push will prompt the Duo Mobile App to send a request
  2. If you type sms, you will receive a code via text that is valid for 10 minutes
  3. If type code, the Duo Mobile App will generate a code that is valid for 30 seconds and can be seen by selecting the three vertical dots in the app
  4. For both sms and code once you receive the code, submit the login information again, but this time use the SMS code or Duo Code as the Second Password

Set Up X11 Forwarding

In PowerShell run winget install X410

https://x410.dev/cookbook/built-in-ssh-x11-forwarding-in-powershell-or-windows-command-prompt/

run X410 by running the command x410.exe
Now run $env:DISPLAY="127.0.0.1:0.0"
Run echo $env:DISPLAY you should see 127.0.0.1:0.0
You may also choose to make DISPLAY a permanent environment variable in your windows machine. To do so, run the command setx DISPLAY "127.0.0.1:0.0"
1. After running setx you will need to close and reopen PowerShell

SSH into SUNNY

Open PowerShell, Visual Studio Code, Sublime or other ssh-capable code editor
In PowerShell run the command ssh -Y <unityID>@SUNNY.ece.ncsu.edu where <unityID> is your NCSU Unity ID
Enter your password and hit Enter (Note: the characters you type will not show up in the console)

note

You can end your remote session without disconnecting from the VPN by typing exit and hitting Enter

Check your access to NFS Lockers

Check that you have access to all of the storage lockers by running the following commands
1. ls /mnt/research-projects/s/screberg/longterm_images
2. ls /mnt/research-projects/s/screberg/longterm_images2
3. ls /mnt/research-projects/s/screberg/GROW_DATA
4. ls /mnt/research-projects/r/raatwell/longterm_images3
If you do not have access to one or more of the lockers, contact mkutugata@ncsu.edu

Install `miniconda` on SUNNY

Download miniconda by running wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
Install miniconda using the command bash ~/Miniconda3-latest-Linux-x86_64.sh
Press Enter to review Anaconda’s Terms of Service (TOS).
Then press and hold Enter to scroll (this will take a little while as the TOS is long)
Once you see Please answer 'yes' or 'no':' Enter yes to agree to the TOS.
Hit Enter to confirm the installation location to your home SUNNY folder
Type yes to set up conda to automatically initialize whenever you open a new shell
Run the command source ~/.bashrc to refresh your terminal and apply the new settings
Test your installation by running conda list. If conda has been installed correctly, a list of installed packages appears.

All information that follows assumes that you are actively logged into SUNNY remotely using ssh. See Setting Up your Machine for instructions on how to do so.

Setting Up the Pipeline

Clone the GitHub repository

In your remote SUNNY session, run the command git clone https://github.com/precision-sustainable-ag/SemiF-Preprocessing.git to clone the SemiF-Preprocessing repository to your SUNNY account
To check if the repository was cloned, you can run the command ls to list everything in your current directory

Navigate to the Preprocessing Directory

Use the command cd SemiF-Preprocessing to change directories

Set Up your `conda` environment in SUNNY

Run the command conda env create -f ./environment.yaml
Then run the command conda activate semif_prep
Install PiDNG by running pip install git+https://github.com/schoolpost/PiDNG.git
Manually install RawTherapee CLI by running the command scripts/validate_rawtherapee.sh.
The conda environment you are in is shown in parentheses before your user <unityID>@sunny:
1. For example, if the base environment is active it will appear as such: (base) <unityID>@sunny:
2. If the semif_prep environment is active it will appear as (semif_prep) <unityID>@sunny:
To change your active environment run the command conda activate <EnvironmentName>
1. To update your environment after making changes to the environment.yaml file, run:
  conda env update --file environment.yaml

Running the Pipeline

In your SUNNY session, change directories to SemiF-Preprocessing using the command cd SemiF-Preprocessing
Make sure that semif_prep is the active conda environment by running conda activate semif_prep
Edit conf/config.yaml by running nano conf/config.yaml
1. batch_id: batch you want to preprocess
2. tasks/copy_from_lockers: copy raw files from LTS to ./data/<lts_location>/semifield-upload/<batch_id> (comment out if already downloaded)
3. tasks/raw2jpg: convert local raw files to jpegs stored in LTS: <lts_location>/semifield-developed-images/<batch_id>
4. tasks/raw2png and tasks/png2jpg can be left commented out unless you need to run these separately
5. raw2png/remove_raws: delete local raw files when each file is converted to pngs
6. raw2jpg/remove_pngs: delete local png files when each png is converted to jpeg
When you have finished editing, hit Ctrl and X at the same time
Type Y to save the newly edited file
Hit Enter to save the newly edited file as config.yaml
Run the command python main.py to execute the main.py python script in the SemiF-Preprocessing directory

.
├── conf
│   ├── ccm     # config files to define color correction matrix (ccm)
│   ├── config.yaml
│   ├── hydra
│   └── paths
├── data
│   └── semifield-utils
│       └── image_development
│           ├── color_matrix        # ccm files saved as numpy arrays 
│           └── dev_profiles        # rawtherapee pp3 profile
├── main.py                         # main entry point for omegaconf
├── scripts
│   └── validate_rawtherapee.sh     # script to validate rawtherapee installation
└── src
    ├── archive                     # archived code for future reference
    ├── raw2jpg.py                  # convert downloaded raw image to jpeg (raw2png + png2jpg processing)
    ├── png2jpg.py                  # convert png files to jpeg
    ├── raw2png.py                  # convert raw images to png
    └── utils
        ├── copy_from_lockers.py    # copy raw files from NFS to local storage
        ├── calculate_ccm.py        # utils to calculate and save ccm based on yaml config 
        ├── debayer.py              # demosaic image to manually record colorchecker values
        ├── preprocess.py           # collection static methods for image preprocessing
        └── utils.py                # common util functions

to end your ssh session type exit

Mobaxterm is a forwarding program - we dont need this for the svcam (we might need this TBD)

maybe get mobaxterm to view and validate images

try and run batch NC_2025-02-21

Need documentation on running the pipeline

be careful as lts is a shared long term storage, do not delete anything

Requirements:

access to SUNNY
miniconda installed https://www.anaconda.com/docs/getting-started/miniconda/install#macos-linux-installation
access to NFS locksers (longterm_images2, longterm_images, GROW_DATA)

Pre-Requisites

Setting Up Your Machine

SSH Setup

PowerShell

Visual Studio Code

Setup Remote Access to SUNNY

Login to NCSU’s VPN

Set Up X11 Forwarding

SSH into SUNNY

Check your access to NFS Lockers

Install `miniconda` on SUNNY

Setting Up the Pipeline

Clone the GitHub repository

Navigate to the Preprocessing Directory

Set Up your `conda` environment in SUNNY

Running the Pipeline

Requirements:

Getting setup

Running the pipeline

Manually inspecting images

Transferring data to LTS if images pass inspection

Creating issues in Github if images fail inspection

Pre-Requisites

Setting Up Your Machine

SSH Setup

PowerShell

Visual Studio Code

Setup Remote Access to SUNNY

Login to NCSU’s VPN

Set Up X11 Forwarding

SSH into SUNNY

Check your access to NFS Lockers

Install miniconda on SUNNY

Setting Up the Pipeline

Clone the GitHub repository

Navigate to the Preprocessing Directory

Set Up your conda environment in SUNNY

Running the Pipeline

Requirements:

Getting setup

Running the pipeline

Manually inspecting images

Transferring data to LTS if images pass inspection

Creating issues in Github if images fail inspection

Install `miniconda` on SUNNY

Set Up your `conda` environment in SUNNY