Menu Bar

View

• Zoom In: Increases the magnification factor of the image.
• Zoom Out: Decreases the magnification factor of the image.
• Red/Green: If this option is set, the two images are displayed in magenta and green, respectively. If both images match perfectly, a gray image is obtained. Color images are first converted to grayscale and then transformed to magenta and green.
• Checkerboard: If this option is set, one image is displayed in the white parts of a checkerboard pattern while the other image is displayed in the black parts. The size of the parts can be adjusted in the Options dialog (see below).
• Average: If this option is set the, two images are averaged or blended in the viewer window.
• Invert: If this option is set, the lower image is inverted. The inverted image is then blended with the upper image. If both images match perfectly, a constant gray image is obtained. This option is the default.
• Transform parameters: If enabled, an additional toolbar is added which displays the transformation and flip state of the currently editable slice.
• XZ view: If enabled, displays a xz view of the data (i.e., aligned on the y axis) and a toolbar for controlling the displayed slice number and its zoom factor.
• YZ view: If enabled, displays a yz view of the data (i.e., aligned on the x axis) and a toolbar for controlling the displayed slice number and its zoom factor.

  If either or both of these views are enabled, an additional toolbar is displayed. Two text fields allow you to specify the min and max Z values of interest. Activate the Auto update toggle for the orthogonal views to be automatically updated each time you change a slice in the main align window. Press the Update button to request an explicit update.

Options

• Undo: This menu provides an undo feature that undoes the last operation. Successive invocation of Undo is possible, allowing several operations to be undone. Each operation made manually or automatically (translation, rotation, automatic alignment, landmarks editing) can be undone. The maximum number of operations that can be undone is 100.
• Redo: This menu provides a redo feature that redoes the last undone operation. Successive invocation of Redo is possible after several undo operations.
• Reset all: Set the translations on X and Y and the rotation to 0 for all slices.
• Reset: Set the translations on X and Y and the rotation to 0 for the current slice (the slice that is currently editable).
• Transform all: This is a toggle option. If the option is ON, the transformations made for the editable slice are also made for all slices above when the editable slice is the upper one. That means that the existing alignment of all other slices not involved in the current transformation will be preserved.
• Fix reference: This option decides if a certain slice is used as a reference slice during the entire alignment procedure. The reference slice is marked by a red tag in the Slice slider. The other slices will be aligned according to this slice.
• Read transformation: Reads the AlignTransform and AlignPoints parameters from the input data and sets the transformation values accordingly. This is useful to return to a previous saved alignment. This can be seen as another kind of reset or undo.
• Save transformation: Save the actual translations, rotations, and landmarks as parameters in the input data.

Align

• Gravity centers: Align gravity centers and principal axes
• Gray values: Least squares algorithm based on gray values
• Landmarks: Align user-defined sets of landmarks
• Edge detection: Align the outer bounds of the objects

In addition, this menu allows you to specify which slices to align, as follows:

• Align current pair: Align just the current two slices
• Align all slices: Align all slices

Finally, the Options button of the Align menu opens a tabbed dialog window allowing you to define a variety of alignment-related options. Each tab will be handled separately below. The tabs are: General, Output, View, Edge Detection, and Least Squares.

Align/Options/General

• Allow rotation in automatic alignment: If this option is checked, rotations are considered during the alignment process. By default, this option is checked. Translations are always considered.

Align/Options/Output

Figure 1: The Output options window.

The following parameters can be set using this dialog:

• Automatic size: If this feature is activated, a minimal image size is calculated which contains all pixels of all slices. The dimensions and the position of the new bounding box are set such that all slices fit in this new bounding box.
• Image width: Sets the width of the output image.
• Image height: Sets the height of the output image.
• Show borders: Shows the new image borders of the output image in the Slice Aligner.
• Background color: The user can set the background color for the output image. The color can be set as a gray value or as an RGB color, depending on the type of the input image. The current background color is depicted in the box to the right.

Align/Options/View

• Size of checkerboard: Adjusts the size of the pattern if view mode is set to checkerboard.

Align/Options/Edge Detection

Figure 2: The Edge detection options window.

The parameters for the Edge detection algorithm can be set using this dialog:

• Matrix dimension: This is the dimension of the matrix which scans the image and decides whether a pixel belongs to the object or the background.
• Matrix percentage: This percentage defines the amount of pixels in the surrounding of a pixel which have to belong to the object to indicate that this pixel belongs likewise to the object. The considered surrounding is defined by the matrix dimension.
• Matrix rastering: This indicates if the surrounding should be rasterized. If so, the alignment is done in a shorter time.
• Image rastering This decides if the image should be downsampled during the alignment procedure. This speeds up the alignment.
• Flipping: This option indicates if a possible flipping of the images should be considered.
• Calculate threshold: This activates an automatic calculation of the threshold.
• Set threshold: This allows the threshold to be set manually.
• Background property: This must be set to describe the background of the image. According to this setting, the threshold is interpreted as an upper or a lower boundary of gray values which separate the object from the background.
• Angle stepsize: This sets the size of the search algorithm's angular steps (in degrees). A higher value speeds up alignment at the cost of accuracy.

Align/Options/Least Squares

Figure 3: The Least squares options window.

The following parameters for the gray values algorithm can be set using this dialog:

• Max number of iterations: This is only intended to prevent an infinite loop.
• Step size for translations: The step size used to search for better positions in X and Y. A small step size, though in general may lead to more accurate results, also slows down alignment.
• Step size for rotations: The step size used for rotations.
• Scale factor: During the alignment procedure, the image can be downsampled for the first iterationsteps so that a better result can be reached in a shorter time.

More information about the align methods and the appropriate settings for the methods can be found in the alignment section.

Landmarks

• Add: set the shape of the mouse cursor to a black landmark and clicking on the slice a new landmark is created.
• Remove: delete the selected landmark. This menu item is active only when a landmark is selected and the number of landmarks is greater than 3.