[SVN revision 22524]
FIX (#3619): Adjust number of points in SetControlPoint()

[SVN revision 22470]
DOC (#3619): Added class documentation

Updated StateMachine.xml

Implemented first version of PlanarCross. The PlanarFigureInteractor and the corresponding state machine needed to be adjusted. This also affects the behavior of other PlanarFigure classes.

[SVN revision 22432]
ENH (#3619): Spatially constrain control points by image bounds

Editing of PlanarCross does not work correctly due to missing state in StateMachine: REOPENED

Added feature calculation, Class documentation, code clean-up

[SVN revision 22355]
ENH (#3619): Basic version of PlanarCross. Small adaptations of PlanarFigureInteractor and StateMachine and other PlanarFigure sub-classes. A few code improvements.

Implemented spatial constraints so that control points remain within image bounds

[SVN revision 22468]
FIX (#3619): Make GetNumberOfFeatures() virtual so that it may be overwritten in sub-classes

[SVN revision 22467]
ENH (#3619): Calculation of both longest and short axis diametersCs; correct application of both image boundary and cross spatial constraints to control points

[SVN revision 22492]
ENH (#3619): Update PlanarFigureInteractor state machine with new state for PlanarFigure reset; update BlueBerry state machine with core/code/interactions state machine

The derivation from itk::Object does not seem necessary.

[SVN revision 21950]
FIX (#3464): Further clean-up of GUI (remove unused elements)

[SVN revision 25674]
FIX (#3463): Under Win-XP, don't use transparency for overlays but paint black background

[SVN revision 21949]
FIX (#3464): Remove legacy methods from QmitkImageStatistics

Issue addressed in T3493, fixed with r22205. Closing bug.

THe methode ProjectPointOntoPlane() internally calculates the distance of a point to the plane. The signed distance is required here, but the unsigned distance was used. Changed this to use signed distance.

[SVN revision 21750]
FIX (#3409): Fixed calculation of point projection and extended test case with a few more points

Fixed with r21572 (IsExclusiveFunctionality() returns false).

[SVN revision 21802]
FIX (#3334): Initialize QItemModel correctly (by setting row AND column count) to make the histogram widget work again with new Qt versions

A small update: this seems to be related to the configuration of VTK and/or Qt.

[SVN revision 21189]
ENH (#3276): Add display of image/segmentation volume to ImageStatistics

Dear Xavier,

Dear Xavier,

(In reply to comment #2)

Created an attachment (id=443) [details]
Unified patch

Modified the default values of top and frontside to be backwards compatible.

Could you please explain me what's the purpose of this code?
mitk::SlicedGeometry3D::InitializePlanes( ):

bool flipped = (top == false);

if ( frontside == false )
{
  flipped = !flipped;
}
if ( planeorientation == PlaneGeometry::Frontal )
{
  flipped = !flipped;
}

(In reply to comment #6)

In the case of this particular image, the extents are 2.048 though, and the
method does not exit. Since the geometry information does not exist yet, the
reslicer returns a 0-dimensional data, and a warning happens, as mentioned by
Matthias.

(In reply to comment #8)

The extents in the first run of GenerateData are computed through the norm of
the axis vectors in index coordinate (lines 403-406). There are many operation
involved:

• Transformation of the vectors in world coordinates to index coordinates,

which involves a few matrix transformations;

• Computation of the norm from the transformed vectors.

As I mentioned before, the other images have values smaller or equal to 2.0 for
the extents in the first run, and the method returns without further processing
(lines 506-509). I've watched the extents of several images and they ranged
between 1.2 and 2.0.

Since these vectors represent the extents in index coordinates, I guess that
float numbers are not allowed. These extents are also used to compute the mm
per pixel values (lines 415-416), so they are definitely the size in pixels,
and I can imagine that having float values for them does not make sense.

This looks like a nice feature but a quite dramatic change in how MITK interaction works. Maybe this could be done optionally? Are there any reasons why we should already have it for 1-0?

Features like this should become possible with the planned extension to the general interaction framework. Won't be fixed here.

This has been solved in T2311.

[SVN revision 21201]
ENH (#3207): Move QmitkImageStatistics to public mitk

The volume of an image (region) analyzed by QmitkImageStatistics is displayed in the statistics data list since r21189 (see T3276).

(In reply to comment #1)

We tried a "tiny" image with dimensions 3x3x1 and spacing 0.5x0.5x0.5 and
Pic3D; MITK doesnt crash and interactions work.

Please provide the original images, where the crash occured.

Setting bug to fixed.

Additional requests:

• histogram range should remain constant in +t datasets
• Add range slider for histogram
• Improve interaction (only x-zoom by default)
• extend class documentation

Other related fixes that were applied with the above commit:

[SVN revision 20822]
FIX (#3155): Changed interface of PlanarFigure to allow name change of features during run-time (relevant when PlanarFigure changes its type at run-time through parameter change)

[SVN revision 21573]
ENH (#3098): Refuse images with more than one component for statistics calculation instead of specific types (this is done in QmitkImageStatisticsView, see r21572 and T3067)

(In reply to comment #9)

Mathias, could you check for crashes in the statistics tool? I will attach
example data sets.

Adding block on T3207 (make ImageStatistics open source)

ImageStatisticsCalculator will be moved to open source for MITK-1-0 (T3207)

Cannot reproduce this issue any more with the current SVN checkout. Might be due to the recent changes in the way PlanarFigures are initialized.

How about

(In reply to comment #8)

Reenabled it on my local system and it seems to work fine, what exactly was the
problem?

First version works nicely, but a line diagram widget should be used instead of bar chart widget for representing line profiles.

[SVN revision 21730]
FIX (#3067): Disable line profile widget since it is not yet ready-to-use

[SVN revision 21572]
ENH (#3067): added rudimentary version of new QmitkVtkLineProfileWidget

Tested for a couple of applications:

[SVN revision 20548]
FIX (#3064): Prevent reentrant calls to UpdateStatistics() invoked by Qt piped update requests

[SVN revision 21145]
FIX (#3064): Replace explicit processing of Qt process loop by immediate update of render windows to avoid re-entrance issues

The reentrance problem was observed in QmitkImageStatistics, which is not (yet) open source.

[SVN revision 21136]
FIX (#3064): Revert changes of r20548 (workaround for QmitkProgressBar re-entrance problem); problem will be solved globally

(In reply to comment #3)

Source files including ProgressBar

Solution to be discussed.

[SVN revision 20405]
FIX (#3058): Proper clean-up of progress listeners in all cases to avoid accumulation of events

Currently, the method mitk::VtkPropRenderer::WriteSimpleText() is used to print text annotations for PlanarFigures. The method internally uses VTK text printing, which is displayed as overlay in MITK by default.

Seems to work now. Closing bug.

Bug has been fixed. A related problem occurs when moving the (rotation) mouse
cursor along an plane intersection line beyond the render window border. The
cursor symbol (wrongly) remains in rotation mode and is only reset when
re-entering the renderwindow at a different location.

Requesting core modification...

[SVN revision 20437]
FIX (#3047): Explicit reset of mouse cursor when slice rotation mode is left

Confirmed. Setting to CLOSED.

This was due to T3011, where no initialization event was issued after finishing placement of PlanarPolygons. Has been fixed. Marking bug as duplicate.

Appears to work correctly now. CLosing bug.

[SVN revision 20330]
FIX (#3039): Returning absolute value of area to avoid negative areas due to which can occur in used algorithm

Confirmed. Setting to CLOSED.

"Area" is still shown when path is created (Circumference = 0.0, Area = 0.0).

Fixed with revision 20420. See T2655. Closing bug.

Confirmed. Setting to CLOSED.

[SVN revision 20320]
FIX (#3031): Now issuing itk::EndEvent at the end of an interaction and setting modified flag of PlanarFigure