Skip to Content

different distortion angles

1 reply [Last post]
Alok
Offline
Joined: 22 Apr 2017

Hi,

I am trying to estimate mag_distortion. I collected 20 micrographs from the gold grid in the super resolution mode. Each micrograph is of 50 frames, frames were aligned and summed. Gain reference correction was applied too. However, I get different values of Distortion angle for different combination of images used to calculate mag_distortion. e.g.

When Images 1-11 are used for the calculation, output is:

The following distortion parameters were found :-

Distortion Angle = 41.2
Major Scale = 1.006
Minor Scale = 0.994

Stretch only parameters would be as follows :-

Distortion Angle = 41.2
Major Scale = 1.012
Minor Scale = 1.000
Corrected Pixel Size = 0.746

The Total Distortion = 1.20%

For Images 11-20 :

The following distortion parameters were found :-

Distortion Angle = 44.4
Major Scale = 1.006
Minor Scale = 0.994

Stretch only parameters would be as follows :-

Distortion Angle = 44.4
Major Scale = 1.012
Minor Scale = 1.000
Corrected Pixel Size = 0.746

The Total Distortion = 1.15%

For Images 1-20:

Distortion Angle = 42.4
Major Scale = 1.006
Minor Scale = 0.994

Stretch only parameters would be as follows :-

Distortion Angle = 42.4
Major Scale = 1.012
Minor Scale = 1.000
Corrected Pixel Size = 0.746

The Total Distortion = 1.15%

Is this common or I am doing something wrong? If the differences are common, how should I choose the final values.

Also, for mag distortion correction, should we use distortion parameter or stretch only parameters (perhaps distortion parameters)?

Thanks so much !

Alok

timgrant
Offline
Joined: 16 Jul 2010
Hi, I assume the 20

Hi,

I assume the 20 micrographs are all taken of different areas of gold? I assume what is going on here, is that the difference is just "noise" on top of the correct answer. In this case, the distortion is relatively small, and a difference of 3 degrees will have almost no effect.

I would use whatever value you get when you run all 20 together as your actual value.

Cheers,

Tim