Event contours are drawn at wrong position for large field of views

[felix-r]

If one measures with ROIs larger than 250 pixels in width, the contours will be only drawn into the first 250 pixels of the image. This might cause funny effects like shown in the image below.

[paulmueller]

Could it be that these are the first 256 pixels of the image? In that case, this might be a problem in ShapeIn, storing the contour coordinates as uint8 @phidahl?

We could implement a workaround for the tdms file format, if this is critical (e.g. if you need to compute ~emoduli~ brightness or volume).

[phidahl]

Check the txt files where the contours are. it should be 32bit and then stored in an asci file.

Von meinem Telefon gesendet

Am 25.07.2017 um 18:14 schrieb Paul Müller notifications@github.com:

Could it be that these are the first 256 pixels of the image? In that case, this might be a problem in ShapeIn, storing the contour coordinates as uint8 @phidahl?

We could implement a workaround for the tdms file format, if this is critical (e.g. if you need to compute emoduli).

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.

[paulmueller]

@felix-r Could you please upload the "M*_contours.txt" file of that particular measurement here or point us to an internal location where to find your data?

[felix-r]

This is the contours-.txt-file for the example above.

M14_0.024000ul_s_contours.txt

[paulmueller]

Ok, it's not a ShapeIn bug. I would need the full measurement to debug this (including the avi file). Anywhere on the group drive is fine.

[felix-r]

It's here: \biotec-files\guck_rtdc\experiments_16_17\20170725_Ruchi_PAA_Beads\PAA_220517_S0_3(1)\Online

The example measurement is M14

[paulmueller]

This was a bug in dclab (see commit message above). Please try the latest build: https://ci.appveyor.com/project/paulmueller/ShapeOut/build/job/khksjp01br3x2ib4/artifacts