PixInsight Module
Only images taken using monochrome sensors are supported at the moment. CFA Drizzle integration should be used for OSC images to get good results.
Starting from PI version 1.9 Lockhart DeepSNR has a PI repository and this is now the easiest way to install it.
Please make sure you have PI 1.9 Lockhart installed if you want to use the repository!
You will need to add two new repositories to PI to make DeepSNR work:
https://pixinsight.deepsnrastro.com/
Optionally (see notes below):
https://pixinsight.deepsnrastro.com/tensorflow/
Notes on installing Tensorflow (TF) libraries:
If you have StarNet installed, you do not need to install Tensorflow libraries again – you can skip the second URL.
If (and only if) you prefer to install the old way, you can download below:
Latest Release (1.0.3)
PixInsight module files (PI 1.9 Lockhart only):
Notes for MacOS: installation procedure is a bit different from before. The files should be copied into MacOS subfolder (not bin as before), so when you go to Install Modules in PI, you will need to change default search path from bin to MacOS as well.
Windows: DeepSNR_win_1.0.3-0036_TF_x64_pixinsight.zip
MacOS: DeepSNR_MacOS_1.0.3-0036_TF_x64_pixinsight.zip
Linux: DeepSNR_linux_1.0.3-0036_TF_x64_pixinsight.zip
Tensorflow library files:
Install if you don’t have any in your PI installation already. If you have StarNet or older version of DeepSNR you can skip these.
Windows: DeepSNR_win_TF_libs_x64_libs.zip
MacOS: DeepSNR_MacOS_TF_libs_x64_libs.zip
Linux: DeepSNR_linux_TF_libs_x64_libs.zip
CLI Tool
Coming soon-ish