Image Integration options not deactivated

[georg.viehoever]

When ImageIntegration opens for the first time, the "Clip low range" and "Clip high range" buttons are active even when "No Rejection" is selected. See screenshot. I think they should be deactivated on "No Rejection". As far as I could see, with "No Rejection" Clip low range" and "clip high range" have no effect.

Georg

[Juan Conejero]

Hi Georg,

There is no bug here. Pixel rejection and range rejection are two different, actually unrelated concepts. You can apply range rejection without selecting any pixel rejection algorithm (that is, with "No rejection" selected), and it will reject all pixels outside the range specified by the range low and range high parameters if the clip low range and clip high range options are selected, respectively.

I admit that the distribution of controls on the II interface is a bit difficult to understand in this regard. The fact that the range rejection controls are located in the same interface sections as the pixel rejection controls (Pixel Rejection (1) and Pixel Rejection (2)) can lead to this confusion. Since I have to publish a new version of the ImageIntegration module (with improved parallel processing), I'll try to improve the II interface to separate range and pixel rejections visually.

[georg.viehoever]

In this case, something else is wrong here:

I am using 4 panels aligned using StarAlignment (actually part of the material provided by Cat/Steve). I am using "Average" as combination method, and I am using "Clip low range" not to use the black background.

The result (and the ImageIntegration setup) can be seen in the attached screenshot. I expected the central parts (2-4 contributing pixels) to have basically the same brightness as the areas where only one pixel contributes to the average. But this does not happen.

Georg

[Juan Conejero]

Hi Georg,

I'd need to take a look at the images in order to understand what's going on here (can you upload them? with Steve's permission naturally). Anyway, use Combination=Maximum and then you won't need any rejection. This is much easier to do with PixelMath, by the way:

max( a, b, c, d )

will do the trick much faster.

[georg.viehoever]

Hi Juan,

I currently dont have the network bandwidth to upload the images. But there is nothing special about them: plain fits files generated by StarAlignment. And the background that should be ignored comes from StarAlignment, not from anywhere else. I have the impression (and the rejection maps seem to confirm it) that the zero pixels are not properly recognized. See screenshot.

Georg

PS: Thanks for the workaround. I am posting this report just because I think that ImageIntegration is not working properly in this case.

[georg.viehoever]

This problem seems to be fixed as well with the version released of the module released in http://pixinsight.com/forum/index.php?topic=3394.0;topicseen#msg23356 .

  Georg