Figure 11.3: Define your display preferences here
Outline color. Push the button to open a dialog where you select the color to be used for the outlines of the measured objects.
Bottom color. Push the button to open a dialog where you select the color to be used for the bottom of each building if the project is set to include the building base.
Fill color. Push the button to open a dialog where you select the color to be used for the interior of the measured objects. The control is only enabled if the Fill polygons check box is checked.
Fill polygons. Check this check box if you want the interior of measured objects to be colorized. The color is blended with the digital image, so you can still see the image. (example)
Transparency. Choose the amount of transparency in percent. The control is only enabled if the Fill polygons check box is checked.
Line width. Enter a positive floating point value, defining the displayed width of the object boundaries.
Show file names. Show or hide the image file name in the top left corner of the two image windows. The file name including file extension but excluding the full path is displayed.
Interpolation method for image zoom. Select which method to use when an image is displayed with a zoom factor larger than 100 percent. Interpolation is always made on the original bit depth (e.g. 11-bit if Ikonos). Interpolation does not change the stored image data, only how the image is displayed.
Center on last object. If checked, the image will be automatically moved so that the last measured object is in the center of the image window.
| Interpolation method | Example (4x zoom) |
| Nearest neighbour. Using nearest neighbour, the original pixel values are just duplicated. The result tends to be blocky, but the method is very fast. |
 |
| Bilinear interpolation. Using bilinear interpolation, a 2 by 2 neighbourhood is used to make a linear estimation of values "inbetween" the original pixel values. The method is faster than cubic convolution, but slower than nearest neighbour. |
 |
| Cubic convolution. Using cubic convolution, a neighbourhodd of 4 by 4 pixels is used to estimate a pixel value "inbetween" the original values. A cubic polynom is fitted in x and y directions. Although 16 pixels are required to estimate a single interpolated value, well known "tricks" (e.g. separability and moving sums) to speed up the interpolation have been employed. Although it is slower than both methods above, the delay should not be distrubing on a modern computer. Cubic convolution increases contrast at edges. You may want to decrease your contrast settings to achieve best viewing conditions. |
 |