SwisTrack/Working with the Display
One of the powerful features of SwisTrack is its real-time display, which allows you to watch the output of every component. Some components even allow you to check out input images or intermediate processing steps.
Configuring the Display
When you click on a component, the display in the main window automatically switches to the output of the selected component. However, you can at any time switch to another display by clicking on the yellow title. Most components provide one display (the output), but some even provide several different displays. (Very few components actually don't provide any display. This is the case for cameras that just receive a stream of data which first has to be translated into a real image, for example.)
If your image is too big to be displayed on the screen, you can drag it with the mouse to inspect the hidden parts. You can also zoom out by clicking with the right mouse button on the image and selecting the appropriate zoom level.
Working with Multiple Displays
Sometimes, it is handy to see several processing steps at the same time. To do that, choose New Display from the View menu. The window that appears has the same functionality as the main display, but doesn't change the display when you click on a component.
Saving Displayed Images
A right-click on a display provides you two possibilities for storing the image:
- Save displayed image as ... stores the image as it is displayed on the screen, i.e. with the chosen zoom level. The image is stored as a color image, no matter what type the original image was. All drawings (particles, tracks, text, ...) are retained.
- Save original image as ... stores the image as it was provided by the component, in the format provided by the component. Note that tracks and particles will not appear on that image.
The black bar below the display indicates status information, such as the frame number of the size, about the displayed image.
As a side note, the displayed image is not always the last processed image. The status information shown here belong to the displayed image.
Display Speed Considerations
Displaying an image on the screen requires some processing and therefore steals some CPU time. If you want to get the maximum performance out of your system, and don't need any display, then choose No display (maximum speed) from the list.
Displaying an image on the screen is done in two steps:
- The component makes a copy of the image it wants to show. This time adds to the processing time (step duration) of the component.
- Once the frame is completely processed, this copy is scaled, converted into a screen image and finally rendered on the screen. This usually takes up much more time than simply copying the image, but does not add to the processing time of the component.
Giving exact numbers is impossible here, as this heavily depends on your hardware and scales almost linearly with the number of pixels of the image. In a simple scenario with not a lot of image processing, however, displaying the image accounts for a significant amount of the whole processing time.