OpenScan2
Overview:
The Raspberry Pi + OpenScan Pi Shield can be used to control two independent stepper motors and a variety of different cameras (Pi Camera, various Arducams, DSLR - via GPhoto and external Cameras like Smartphone and others). The mechanism can be used in various forms (see for example OpenScan Classic or OpenScan Mini and it could be easily adapted to be used as a camera slider or in other mechanisms. After sticking to my original code for way to long, I have spent a lot of time rewriting and re-organizing the whole code/structure/documentation and started this new repository. This should make it much easier to collaborate, implement future updates and continue exploring the wonderful world of 3d scanning :)
I have pre-compiled a working Raspbian Image (2022-04-22), that can be downloaded from Google Drive (1.5GB) (working flawlessly on Raspberry Pi 3b+. Need to re-install the camera drivers on Raspberry Pi 4!)
Changelog
2022-04-21
- added: timer (ETA) until a routine is done
- added: showing progress, while files are being split (before uploading to OpenScanCloud)
- added: infotexts (FINALLY :)
- added: several stats/device information
- fixed: combining two sets did not delete the smaller set
2022-04-20
- !fixed: pi cameras (v1.3, v2.1 and HQ) finally work and can be simply selected in the settings menu
- !fixed: Raspberry Pi 3B+ and 4 work! (the main limiting factor now is the RAM, where at least 1GB RAM is needed)
- fixed: live preview sometimes did not work. This has been a network speed issue and has been solved by downscaling the image (resolution can be set)
- fixed: it is now possible to delete individual sets.
- fixed: it is now possible to use all LEDs.
- added: Turntable mode (disable the second axis)
- added: Pause scan. You can pause and un-pause the scan by simply pressing the button
- added: second scan pass. When one scan is done, you can immediately run a second pass. This is especially useful, if you want to re-orient the object
- added: auto-timeout. Turn off the ringlight (todo: and motors) after 300 seconds (value can be set)
- added: diskspace warning. When free diskspace drops below a given threshold (4GB by default), a warning message will appear
- changed: new background image, minor design changes
- changed: log file can be easily generated and downloaded by clicking a button (update&info tab)
Main Features:
- use your Raspberry Pi to take very consistent image sets for photogrammetry (other use cases should be possible to ;)
- use the included cloud processing function to automatically create highly detailed 3d models (see OpenScanCloud)
- various supported camera modules:
- standard Raspberry Pi Cameras (v1.3, v2.1 and HQ)
- Arducam IMX 519 (with 16mp and autofocus)
- and many other cameras through the libcamera framework
- DSLRs:
- triggering any camera through an isolated GPIO signal using a modified remote control
Pre-compiled Raspbian image
You can download a pre-compiled Raspbian image from Openscan.eu
Structure of this repository
Many free image datasets with additional information and discussion
- datasets/
- readme.md #todo
Several manuals on how to build/use/adjust the OpenScan devices, firmware and electronics
- manual/
- readme.md #todo
- firmware-browserinterface.md #todo
- firmware-backend.md #todo
- object-preparation.md #todo
Providing the update files for the firmware, where the two .json files give a short summary of what changed
- update/
- update.json
- beta/
- main/
Setup
Prerequisites
It is necessary to use a Raspberry Pi with at least 1GB of RAM to run the firmware. This includes the Raspberry Pi model 2B, 3B, 3B+ and all 4B variants. The firmware and following installation guide is optimized for Raspbian Bullseye with Desktop environment, but can also be used headless (Note, that the desktop version comes with some libraries that you will otherwise need to install manually).
Raspi-config
sudo raspi-config
--> performance options --> GPU --> change to 256 Increase GPU memory, which helps processing the captured photos
--> localization settings --> wlan country --> your country Change the WLAN settings, so that device works in your area
Save and reboot now (or later ;)
Folder structure
Create directories
mkdir /home/pi/OpenScan/ /home/pi/OpenScan/scans /home/pi/OpenScan/files /home/pi/OpenScan/settings /home/pi/OpenScan/tmp /home/pi/OpenScan/updates
WLAN settings
Open the settings file
sudo nano /etc/network/interfaces
And add to the end:
auto wlan0
iface wlan0 inet manual
wpa-conf /etc/wpa_supplicant/wpa_supplicant.conf
Samba (optional but highly recommended)
Create a shared network drive, so that you can easily access the Raspberry Pi's filesystem. It enables you to upload custom datasets to the OpenScanCloud through the browser interface. Simply copy a zip file containing your photos to the ../OpenScan/scans directory.
Install the package
sudo apt-get install samba samba-common-bin
Open the configuration file
sudo nano /etc/samba/smb.conf
And change the following lines:
read only = no
create mask = 0775
directory mask = 0775
Add the following line, if you are using Windows:
wins support = yes
And add to the end of the file:
[PiShare]
comment=Raspberry Pi Share
path=/home/pi/
browseable=Yes
writeable=Yes
only guest=no
create mask=0777
directory mask=0777
public=yes
Set a password for user pi (by default I use password raspberry)
sudo smbpasswd -a pi
And add rights:
sudo chmod -R 777 /home/pi/OpenScan
Gphoto
Necessary to controll DSLR cameras connected by USB cable.
sudo apt install libgphoto2-dev gphoto2
sudo pip3 install -v gphoto2
NodeRed
Download and install:
bash <(curl -sL https://raw.githubusercontent.com/node-red/linux-installers/master/deb/update-nodejs-and-nodered)
--> confirm with yes and yes :)
Open node-red daemon file:
sudo nano /lib/systemd/system/nodered.service
add to the section [Unit]
Wants=network.target flask.service
After=flask.service
and change the following lines under [Service] to run nodered as root:
User = root
#Group = pi
Initialize node-red:
sudo node-red admin init
- settings file:
/root/.node-red/settings.js - Security:
No - Project:
No - Flows File Settings:
Enter - Passphrase:
Enter - Theme:
default - Text editor:
default - External Modules:
Yes
Open node-red settings file:
sudo nano /root/.node-red/settings.js
And find, uncomment and change the following parameters:
userDir: '/home/pi/OpenScan/settings/.node-red/',
uiPort: process.env.PORT || 80,
httpAdminRoot: '/editor',
httpStatic: '/home/pi/OpenScan/',
ui: { path: "" },
functionGlobalContext: { // enables and pre-populates the context.global variable
os:require('os'),
path:require('path'),
fs:require('fs')
},
Enable nodered daemon and restart the device:
sudo systemctl enable nodered.service
sudo reboot -h
Open the node-red settings directory:
cd /home/pi/OpenScan/settings/.node-red/
And run the following command to install some additional palettes to node-red
sudo npm i node-red-dashboard && sudo npm i node-red-contrib-python3-function && sudo npm i node-red-node-ui-table
node-red-restart
Libcamera - to run Arducam IMX519 (and other non-Raspberry-Pi camera moduls)
See Arducam.com for more details
Download all necessary files
wget -O install_pivariety_pkgs.sh https://github.com/ArduCAM/Arducam-Pivariety-V4L2-Driver/releases/download/install_script/install_pivariety_pkgs.sh && chmod +x install_pivariety_pkgs.sh
sudo apt update
./install_pivariety_pkgs.sh -p libcamera_dev
./install_pivariety_pkgs.sh -p libcamera_apps
./install_pivariety_pkgs.sh -p imx519_kernel_driver
Download OpenScan Firmware
Custom node red flows (browser interface):
sudo wget https://raw.githubusercontent.com/OpenScanEu/OpenScan2/main/update/main/flows.json -O /home/pi/OpenScan/settings/.node-red/flows.json
Some Python functions used by the firmware:
sudo wget https://raw.githubusercontent.com/OpenScanEu/OpenScan2/main/update/main/OpenScan.py -O /usr/lib/python3/dist-packages/OpenScan.py
A local server providing several functions (flask):
sudo wget https://raw.githubusercontent.com/OpenScanEu/OpenScan2/main/update/main/fla.py -O /home/pi/OpenScan/files/fla.py
Custom config.txt file, which is needed to use different camera moduls (especially IMX519):
sudo wget https://raw.githubusercontent.com/OpenScanEu/OpenScan2/main/update/main/config.txt -O /boot/config.txt
Arducam's camera focus script for the IMX519 sensor:
sudo wget https://raw.githubusercontent.com/OpenScanEu/OpenScan2/main/update/main/Arducam.py -O /usr/lib/python3/dist-packages/Arducam.py
And the OpenScan Logo to have a nice background:
sudo wget https://raw.githubusercontent.com/OpenScanEu/OpenScan2/main/update/files/logo.jpg -O /home/pi/OpenScan/files/logo.jpg
Enable Flask local server:
create and open the service file: sudo nano /lib/systemd/system/flask.service
with the following content:
[Unit]
Description=photo service
After=multi-user.target
[Service]
#ExecStartPre=/bin/sleep 5
ExecStart=/usr/bin/python3 /home/pi/OpenScan/files/fla.py
StandardOutput=inherit
StandardError=inherit
Restart=always
RestartSec=5
User=root
[Install]
WantedBy=multi-user.target
Enable and start the service:
sudo systemctl daemon-reload && sudo systemctl enable flask.service && sudo systemctl start flask.service
others
add to /boot/config.txt to disable display which causes some issues --> WHY???
hdmi_blanking=2
?
and consecutive code piece:
71 Nov 26, 2022
41 Dec 22, 2022
10 Dec 15, 2022
25 Dec 20, 2022
1.7k Dec 19, 2022
8 Nov 4, 2022
27 Dec 24, 2022
2.5k Dec 22, 2022
22.4k Dec 29, 2022
14 Aug 10, 2022
4 Oct 25, 2022
4 Apr 20, 2022
19 Mar 31, 2022
44 Dec 30, 2022
24 Oct 4, 2022
6 Nov 11, 2022
6 Mar 17, 2022
4 Dec 21, 2022
4 Jun 9, 2022