TECHNICAL SUPPORT: FAQs

Auto-Montage Product FAQ

FAQ 1:Why does my Stereo Pair or Anaglyph appear "inside out" when viewed?

The first image captured should be the top (nearest the lens) of the sample, and subsequent images should be captured in order towards the bottom of the sample. To view an Anaglyph image, you will need spectacles with a red filter over your left eye and a green filter over your right eye.

FAQ 2:How many images may I capture in a single sequence?

Up to 255 images may be captured, for which you will probably require our integrated Z-stepper. Processing a data-set of this size will obviously take longer than for a smaller set of images.

FAQ 3:How do I remove unwanted artifacts from the montaged image?

It is important to ensure that the original images are as clear and free of flare as possible before performing the Scan Montage operation, by adjusting the lighting or optical system. Increasing (or occasionally decreasing) the number of source images captured for a particular size of sample may also improve the results.
Experiment with adjusting the montage operation's patch size and filter, as described in FAQ 006 and with the Scan Montage command.
The montaging process occasionally selects unwanted features on a plane which confuse the generated image. This may be corrected manually by using the edit tools (Edit Menu or Edit Toolbar), to "paste in" regions from other image planes.

FAQ 4:Why are my source images entirely out of focus?

Unless the camera has been accurately installed on the microscope tube, an image focused using the eyepieces might not be in focus at the camera. Focus the image at the camera using the live video display on the source image window, ignoring the eyepieces completely.

FAQ 5:Why do my measurements take so long to update?

When the Measurement Table window is displayed (View Measurements command), all measurements are plotted by default. If you have made a lot of measurements, these may be slow to update, even on a highly-specified machine. On the Measurements property page (Tools Options command), choose to plot only those measurements selected on the Measurement Table.

FAQ 6:Why can't I see Stereo Pair images?

The Stereo Pair image requires a certain amount of practice to view. This author personally finds the technique almost impossible to master, and prefers to use the Anaglyph image or the Color Relief image.

FAQ 7:Do I have to save so many files?

When you save a data-set, a number of associated image files are also written to disk, as described in Files Overview. These files are all necessary if you subsequently require to open the data-set again to perform further scan operations. However, each data-set may use a lot of disk-space.
If Prompt to Save Source Files is selected on the Files property page, on saving a data-set you will be prompted to choose whether to keep your source images or discard them. If you do not save the source images to files, you will not subsequently be able to examine the source images or perform a Scan Montage operation. It is also possible to export a single source or result image to an arbitrarily-named image file without saving the whole document, as described in Export Single Image to File, and then close the data-set without saving it. In this case, you will not subsequently be able to use the data-set.

FAQ 8:How do I make the image windows bigger?

Any image window may be resized as follows:

Move the cursor over the border of the window and wait for the "arrow" cursor to change to a "resize" cursor.
Hold down the left mouse button and pick up the border and drag it out to resize the window.
Release the mouse button.

Note that the bottom right-hand corner of the window is easiest to pick up.
Note also that you may resize the source image window even during live capture.

When initially installed, the main window does not fill the entire screen. Image windows are created to fit the (small) available area. The main window may be "maximized" as follows:

Click on the "maximize" icon on the title bar, the middle icon at the top right corner of the main window. Subsequently opened image windows will now be larger than before.

The next time you start the application, the main window will be maximized, and image windows will therefore be larger.
If not required, some toolbars may be switched off (or on) using the Toolbars menu.
The toolbars may be made smaller (or larger) using the size control on the Windows property page.

FAQ 9:Why am I not seeing measurements displayed?

Measurement results may be displayed (on the status bar and on the Measurement Table window) as uncalibrated pixel and plane values or as calibrated values. The calibration for the X Y dimensions is separate to the calibration for the Z dimension, and they interact as follows:

If neither XY or Z are calibrated: XY and Z measurement fields will all be displayed in terms of raw pixel co-ordinates and raw plane numbers. Certain other measurement fields are not displayed (as they can only be calculated from calibrated values for both XY and Z).
If XY is calibrated but Z is not calibrated: only XY measurement fields will be displayed (in calibrated units). Z measurements and certain other measurement fields are not displayed.
If Z is calibrated but XY is not calibrated: only Z measurement fields will be displayed (in calibrated units). XY measurements and certain other measurement fields are not displayed.
If XY is calibrated and Z is calibrated: all measurement fields will be displayed in calibrated units.

FAQ 10:What does the Patch Size do?

The patch size is a parameter to the Scan Montage operation, and corresponds to the size of detail you are attempting to focus on. It is difficult to say exactly what patch size will suit a particular image, but some guidelines are:
A small patch size, say 5, tends to be particularly effective on images with very fine detail randomly distributed between layers (eg. a transparent biological sample).
A large patch size, say 40, should be more effective where the sample slopes continuously from one z-plane to the next (eg. a metallurgical surface).
The patch size and other parameters are discussed in some detail with the Scan Montage command.
It is recommended that you experiment with a selection of your own samples to find a scan method and patch size that produces the best results for a typical sample, then use those parameters for all similar samples. You may find it convenient to store these parameter values in the template, which may be reloaded for each new session. See also FAQ 003.

FAQ 11:How do I create several data sets from the same source images?

If you have previously captured a set of source images and saved the data-set, you may then re-open the same source images as a new data-set as follows:

On the Files menu, choose Open.
On the Open dialog, change "Files of Type" to "Image Files".
Navigate to your previously saved source images and select any source image file.
Click Open.

You may now process these source images normally and save the results to a new data-set. The original data-set will not be modified. You may create any number of data-sets from a single set of source images.

FAQ 12:How do I work without an image aquisition board?

If you do not have supported framegrabber hardware, you may still create source image files as follows:

Using any other application (perhaps on a completely different system), capture a sequence of source images and save them to suitable filenames.

You may then open these source images as a new data-set as follows:

On the Files menu, choose Open.
On the Open dialog, change "Files of Type" to "Image Files".
Navigate to your previously saved source images and select any source image file.
Click Open.

You may now process these source images normally and save the results to a new data-set.

FAQ 13:How accurate are the measurements?

Measurements in the XY dimensions should be accurate to within the calibrated XY size of a single pixel (as displayed on the Calibrate property sheet), assuming that the optical system is free of distortion and that the initial calibration was made accurately.
Measurements in the Z dimension made on a Blended Depth montage image should normally be accurate to within half of the calibrated plane gap (as displayed on the Calibrate property sheet), assuming that an accurately-calibrated Z-stepper was used during capture. The Z-stepper's own accuracy is to within 2 microns, which may be significant at high magnification.
Measurements in the Z dimension made on a Best Depth montage image are less accurate as the depth map contains only integer values.
Measurements in the Z dimension made on a Weighted Depth montage image are not accurate at all.
However, it is important to realize that artifacts on individual source images may occasionally confuse the montage operation and place a particular result pixel on completely the wrong plane, invalidating any measurement which involves that pixel.

FAQ 14:What is a "feret" and what is a "displacement"?

The X, Y and Z Feret measurements (obtained from Straight Line, Curved Line and Area and Volume measurements) are the absolute distance between the minimum and maximum extents at any point within the measurement. Feret is also sometimes called the "caliper diameter". Ferets always have positive values.

The X, Y and Z Displacement measurements (obtained from Straight Line and Curved Line measurements) are the signed distance between the start and finish points. Displacements might have positive or negative values.

Note that in the illustration, the Y-displacement will have a negative value.

FAQ 15:What are the restrictions on the demonstration version?

This software is protected by a security key, but most functions are available to allow evaluation before purchase. Please see "Demonstration Mode" in the online help.

FAQ 16:Why am I unable to export to Excel?

The ability to export measurements and export line profiles to a Microsoft Excel worksheet for further analysis is a powerful feature. However, it requires a recent version of Excel to be installed, Excel 97 or later.

FAQ 17:What does the message “No free Prysm” mean?

The message “No free Prysm” may be displayed if the Prysm framegrabber support software has not been installed correctly or has become corrupted.
However, if you have previously captured images successfully, it is much more likely that the Prysm card has become dislodged from its PCI slot (particularly if the system has been moved). Open the computer and check that the card is seated correctly.
WARNING: Make sure the computer is disconnected from the mains supply before opening the unit. Additionally, ensure that you observe normal anti-static precautions before performing any work inside your computer. Electronic devices may be destroyed if handled incorrectly. Please contact your distributor for further advice.

FAQ 18:Why does the 3D Model appear slowly on my machine?

The 3D Model uses technology called “OpenGL”. Some graphics cards (“VGA adaptors”, “monitor cards”) offer OpenGL acceleration, which significantly improves the performance of this view.
However, if the graphics card in your PC does not provide OpenGL acceleration, display of the 3D model will be much slower. You can compensate for this by adjusting the Detail control to provide coarser rendering.

FAQ 19:Why does the 3D Model render incorrectly on my machine?

The 3D Model uses technology called “OpenGL”. Unfortunately, we have found that some graphics cards (“ VGA adaptors”, “monitor cards”) which offer OpenGL acceleration do not implement the standard correctly.
The manufacturer of your graphics card may provide updated driver software, check their website for the latest drivers.
If new drivers are not available, open the Control Panel (from “Settings” on the “Start” menu), double-click on the “System” icon, select the “Performance” tab and click “Graphics…”. Reduce the hardware acceleration slider, experimenting with the 3D model to find a setting that provides correct rendering with reasonable performance.
If the software crashes when you attempt to view the 3D model, there may have been a problem installing one of the system components. Please try re-installing from the CD. If this does not solve the problem you must register the component manually. For Windows 95 or Windows 98:

Click the Windows Start button,
choose Programs,
and then click MS-DOS Prompt.
Enter the commands:
cd \windows\system
regsvr32 io3dview.dll

For Windows NT, please contact your distributor.

FAQ 20:Why do I see strange “echos” around images from a stereo microscope?

Stereo microscopes, while often producing excellent source images for this application, may also present particular challenges as the objective moves a considerable distance relative to the sample.
Lighting
If a reflective sample (eg. a metalic object) is lit by a ring-illuminator around the microscope objective, the angle of lighting will change and may cause transient highlights as the microscope is focussed on successive planes. If a transparent or translucent sample (eg. sand grains, glass structures) is lit transmissively, refractive effects may cause moving halos as the microscope is focussed on successive planes.
Fiber-optic “gooseneck” illuminators may offer more consistent lighting in both cases above.
Parallax
The three-dimensional shape of some samples (eg. seashells, mechanical components) may cause some regions to be occluded by other parts of the same sample as the microscope is focussed on successive planes. This is a particular problem where the base of the sample is “undercut” relative to the rest of the sample. Possible solutions include:

Rearrange the sample under the microscope.
Place the sample on a textured surface to provide focussed background detail.
Edit the montage image by hand.

FAQ 21:Why does the 3D Model print incorrectly on my machine?

Surprisingly, printing uses your system’s graphics card (“VGA adaptor”, “monitor card”) to render an image to the printer. Unfortunately, we have found that some graphics cards do not implement OpenGL correctly, which affects printing. Please see FAQ 019.

FAQ 22:What does “Z-stepper not responding” mean?

First check whether the z-stepper motor unit is responding correctly by turning the knob and observing the focus motor. If the motor turns, it is likely that the processor in the control unit, the amplifiers and the motor itself are all probably working correctly. The problem is most likely to be in the computer connection. The z-stepper should be connected to the computer's serial port. Most PCs have 2 ports, called COM1 and COM2, although there is often no labelling on the case to say which is which. To tell the software which port is in use, click "Options" on the "View" menu to present the application options dialog, click the "Z-Stepper" tab and select the appropriate serial port COM1 or COM2. It is suggested that you try both settings, and that you try swapping the serial cable between both connectors on the back of the computer. Serial cables come in 2 variants, "one-to-one" and "null modem". The z-stepper needs a null modem cable, so it is important to use the original cable that came with the equipment. If you suspect that this cable has been lost or damaged, replacements are easily obtainable, but note that a "null modem" cable is NOT the sort you use to connect a computer to a modem.
The following suggestions concern the possible failure of the computer itself:
The computer's operating system may have lost the port configuration (although this is very unlikely unless the motherboard battery has failed). On the start menu, click "Settings" and click "Control Panel" to present the control panel. Double-click "System" to present the system properties dialog, and click the "Device Manager" tab. Navigate down the tree to the "Ports (COM LPT)" branch and check that all devices report themselves as "working normally". Repair should only be attempted by an experienced computer technician.
The computer's internal connections may have become dislodged. Repair should only be attempted by an experienced computer technician, as any static discharge from your body or tools will destroy the exposed computer components.