Rotate3D can reorient an input image arbitrarily in 3D. It can also resample the data to a finer or coarser grid. Quadratic interpolation is used to determine the output values.
A command-line application, rot3d, has many of the same capabilities as Rotate3D.
Overview | Region processing | Angle | Pixel size | Output dimensions | Rotation center | Translation | Fill value | Rolloff distance
Priism | rot3d | Flip | Resample2D
The three values in this field in the main menu are the rotation angles applied to the x'', y', and z coordinate axes. As suggested by the primes, The order in which the rotations are applied is the opposite of the order in which the values are listed in the field: the first rotation is about z and the last rotation is about the rotated x axis.
The angles are assumed to be in degrees. The sign convention is:
When only rotating about the z axis, this sign convention is opposite of that used in Resample2D.
The point which remains fixed during the rotation is set by the rotation center controls in the Options Menu. By default, it is the geometric center of the selected input region.
Overview | Region processing | Angle | Pixel size | Output dimensions | Rotation center | Translation | Fill value | Rolloff distance
This field in the main menu controls the x, y, and z pixel spacings for the output volume. The units are the same as the units on the pixel spacings for the input data. When a new input data set is selected, the pixel spacings in the field are updated to be the same as the input pixel spacings.
Overview | Region processing | Angle | Pixel size | Output dimensions | Rotation center | Translation | Fill value | Rolloff distance
The x, y, and z sizes of the output file are set by two controls in the Options Menu. The first is the Autosize toggle. When it is on, the output size is such that the full volume of the input region is included in the output file. When it is off, you can use the second control, the text field just below the toggle, to directly set the x, y, and z sizes. By default, the Autosize toggle is on.
Overview | Region processing | Angle | Pixel size | Output dimensions | Rotation center | Translation | Fill value | Rolloff distance
The rotation center is the point which remains fixed during the rotation, and, unless a non-zero translation is applied, it is the point at the center of the output volume.
If the input region has only one wavelength and one time point, you can turn the Use center of intensity toggle on in the Options Menu to have the rotation center be located at the center of intensity of the input volume. Otherwise, the x, y, and z coordinates of the rotation center are shown in the text field just below the toggle and can be changed. The coordinate system used has (0, 0, 0) at one corner of a z stack in the input source and (nx, ny, nz) at the opposite corner where nx, ny, and nz are the full dimensions of the input source (i.e. not necessarily the dimensions of the region to process). The lower lefthand corner of the ith z section (i running from 0 to nz - 1) in the input source is at (0, 0, i - 0.5).
Whenever a new region or file is selected, the rotation center is reset to the geometric center of the new region.
Overview | Region processing | Angle | Pixel size | Output dimensions | Rotation center | Translation | Fill value | Rolloff distance
The translation field in the Options Menu sets the x, y, and z displacement of the center of the output volume from the rotation center. The displacements are measured in units of output pixels.
Overview | Region processing | Angle | Pixel size | Output dimensions | Rotation center | Translation | Fill value | Rolloff distance
In the output volume, points which map to back to points outside of the input and more than the rolloff distance from the input volume's edges, are filled with a fixed value. When the Use perimeter average toggle in the Options Menu is on, this fill value is the average of the pixel values from the perimeter of the input volume. When that toggle is off, the value shown just below it is used. The value may be complex; the imaginary component is only used when the output data is complex.
Overview | Region processing | Angle | Pixel size | Output dimensions | Rotation center | Translation | Fill value | Rolloff distance
In the output volume, points which map back to points that are outside of the input volume but which are within the rolloff distance of the input volume's edges have their values calculated from a linearly weighting of the fill value and the value interpolated at the closest point which maps back to a point in the input volume.
The rolloff distance is measured in the same units as are used for the input data's pixel spacing. The interpolation algorithm employed is more efficient (~25%) when the rolloff distance is zero.
Overview | Region processing | Angle | Pixel size | Output dimensions | Rotation center | Translation | Fill value | Rolloff distance