[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Octave-bug-tracker] [bug #52926] [octave-forge] (image) regionprops('pe
From: |
Carnë Draug |
Subject: |
[Octave-bug-tracker] [bug #52926] [octave-forge] (image) regionprops('perimeter') gives unexpected error / 'eccentricicty' inaccurate |
Date: |
Fri, 19 Jan 2018 10:58:39 -0500 (EST) |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Firefox/52.0 |
Follow-up Comment #3, bug #52926 (project octave):
* Throwing an error on the 'perimeter' property with conn=4
The problem is that the Perimeter property must be computed by measuring the
distance around elements using connectivity. That is part of Matlab
documentation. The problem arises when instead of a bw image, we use a
labelled image, or a bwconncomp structure, that was generated with a different
connectivity and would identify different objects with connectivity 8.
I have fixed it in the stable branch now with
http://hg.code.sf.net/p/octave/image/rev/41b32e3648d3
* slightly different results when measuring Perimeter
This is Matlab being backwards incompatible. I went to look into Matlab
Release Notes and seems like they changed it 2014a
> regionprops function uses new algorithm to calculate perimeter
>
> The regionprops function uses a new algorithm to calculate a perimeter, when
used with the 'Perimeter' option. Because of this change, regionprops returns
different results for the perimeter calculations than it did in earlier
releases.
>
>Compatibility Considerations
>
> While the new algorithm used with regionprops returns more accurate
perimeter calculation, you can get the same return value as previous releases
by specifying the 'perimeterold' option.
I have opened a new bug #52933 for it
* slightly different results when measuring eccentricity
This issue was fixed on the image package version 2.6.2. The computation of
Eccentricity is dependent on MajorAxisLength and MinorAxisLength and version
2.6.0 and 2.6.1 had a bug on those. This has been fixed and released.
_______________________________________________________
Reply to this item at:
<http://savannah.gnu.org/bugs/?52926>
_______________________________________________
Message sent via/by Savannah
http://savannah.gnu.org/