OrthoProjection

MLModule
genre Resample
author MeVis Medical Solutions AG
package MeVisLab/Standard
dll MLGeometry1
definition MLGeometry1.def
see also Resample3D, OrthoSwapFlip, SubImageStatistics
keywords max, min, add, mip, stddev, variance, argmax

Purpose

The module OrthoProjection projects a volume dataset along a given dimension using a specific function.

Details

If a dimension is chosen whose extension is 1, nothing is done. (For standard deviation or variance mode, the output will be marked invalid, since a variance estimation needs more than one value.)

For add mode, the image’s minimum and maximum voxel values and the output datatype are adjusted in order to prevent overflows. Unsigned datatypes are preferred in this case.

For the stdDev and var modes, the output image type will be a floating point type, and the image’s minimum is set to zero, but the maximum cannot be properly estimated. So, do not rely on the maximum value in these modes!

Windows

Default Panel

../../../Modules/ML/MLGeometry1/mhelp/Images/Screenshots/OrthoProjection._default.png

Parameter Fields

Visible Fields

Dimension

name: dimension, type: Enum, default: z

Defines the image dimension in which the projection is performed.

Values:

Title Name
x x
y y
z z
c c
t t
u u

Mode

name: mode, type: Enum, default: max

The projection mode determines how the target value is computed from the N source values along the projected dimension. Also, the output type and the stored min/max are affected by some modes, as noted in the table below.

Values:

Title Name Description
max max Maximum of all source values
min min Minimum of all source values
add add Sum of all source values (promoted type)
mean mean Arithmetic mean of all source values, i.e. sum divided by N (type is not changed)
max Pos maxPos Integer voxel coordinate of the maximum value along projected dimension
min Pos minPos Integer voxel coordinate of the minimum value along projected dimension
std Dev stdDev Standard deviation of all source values, i.e. square root of variance (floating point type)
var var Unbiased estimate of the variance of all source values (floating point type)