UPDATE FROM MABULA
I have had a very rough 2 months unfortunately health wise. I was struck 3x in a row with bacterial infections. The second infection occurred  after a routine hospital checkup and I became very sick. I had to rest and take a lot of antibiotics. Once I was recovering and restarting work 1 month ago, I again became sick. The infection was not yet gone, so even more antibiotics and rest was needed. Needless to say, it took a lot of my energy and I needed a lot of rest.
Finally, the infection is really gone and my energy is coming back step-by-step now and I have started work again. I am terribly sorry to have kept you waiting for support. I will address all outstanding questions and e-mails step-by step and will be on the forum daily from today.
APP 2.0.0-beta47 will come soon as well, a lot of work was already completed before I became very ill, so the release is also nearly ready for you. Beta47 will be much faster actually. Many workflows will be more than 2x faster, mosaics can even be 10x faster than before because registration really received a major boost... all compared to beta46. Â Before I release it, I will make sure that everything is working properly and then I will release it.
Hi @mabula-adminÂ
Sorry for the late reply ! I have been busy lately.
As you asked, the complete dataset is uploaded in the folder ''Fred-OutlierRejection-issue''
I've just tried the median integration but don't see any improvement from average integration !
Yes I've always used LNR and also LNC (tested with 2nd et 4th degree with 3 iterations).
Thank you for you assistance
Fred
Hi Fred @fredmt,
I am testing your data now, trying to improve this 😉 Just a question for now, it is clear that the data has many satellite stripes and I am wondering if you have tried adjusting the outlier rejection with the diffraction protection setting as well? Or did you not touch it?
Mabula
Â
Hi Fred @fredmt,
I have played with your data, and you want the diffraction protection to be set to none ! or very high. With 5, all those satellites will survive easily.
I am still testing and will probably improve this behaviour in the next version 😉
Mabula
Hello everyone. If you switch off the automatic mode in tab 6 under Integration and set median, all satellite tracks are removed.
The only disadvantage is that you have to do without multiband blending and Bayerdrizzle. With Bayerdrizzle the stars would have less color fringing.
If you switch from quality to SNR at weights, you still have a better object signal without having to deal with new color gradients (for example through clouds etc.). The SNR setting follows the brightness of the sky background well. So images with a darker background are weighted higher.
Although it is considered a dangerous weighting option in the tooltip.
Integration Average
Integration Median
Greetings Henry.
Hi @minusman,
Thank you very much for your information and help in this issue, I am working on this issue and I can confrim that with median integration, the satellites are cleanly gone. The downside is that median integration is inferior when it comes to noise reducation in stacking when compared the median integration. I am working at the moment on Local Normalization Rejection 2.0 to solve this while still maitaining good SNR in the integrated result.
I have done extensive testing using Fred's data @fredmt and the there are so many satellites in the data that the unrejected pixelstacks remain biased with some residual outliers. And these residual outliers seem to not be removed due to a normalization issue, like you suggested earlier, so I think that is a very good analysis. My solve would be to correct the pixels in the pixelstacks also for dispersion with Local Normalization Rejection 2.0. The current Local Normalization Rejection only corrects for the sky background. Adding dispersion in the correction might solve this. I will soon know if it works 😉
Mabula
