3.1 Java1489-plotting 3d surfaces using java  (Page 5/24)

While the highest elevation is pretty well indicated in the Grayscale plot also, if the central peak were not symmetric in all four quadrants, there mightbe some uncertainty as to the exact location of the highest elevation.

Quantitative estimates are possible

In addition to providing a good overview of the shape of the 3D surface, the Color Shift plot also makes it possible to estimate the elevations of points onthe surface in a quantitative way. For example, in addition to black and white, the yellow and aqua colors are fairly easy to identify on the surface plot andare also fairly easy to identify on the calibration scale. In addition, the yellow and aqua bands on the calibration scale are fairly narrow. By measuringthe locations of these colors on the calibration scale, the elevations of the yellow and aqua areas on the surface plot can be estimated with reasonableaccuracy.

The ranges of the surface elevations colored red, green, and blue can also be estimated but with less certainty.

(Because the green portion of the calibration scale is about twice as wide as the red and blue portions, the level of uncertainty when using thegreen calibration data to estimate the elevation of a point on the surface is about twice the level of uncertainty when using the red or bluecalibration data to estimate the elevation of a point on the surface.)

Color Contour plot

The rightmost image in Figure 1 , which is a Color Contour plot, improves on the ability to provide good quantitative estimates of surface elevations.

If you need to read elevations off the surface plot to a high level of accuracy, the best approach is probably to use a Labeled Numeric Contour plot .. However, such plots are relatively difficult to create. Also, because of the need for the labels to be large enough to read,the space required to display such a plot can sometimes be excessive.

The Color Contour plot is a reasonable compromise between a Labeled Numeric Contour plot and a Color Shift plot. This plotting format provides more accuracyin estimating surface elevations than the Color Shift plot, but doesn't require any more space to display.

Similar to a contour map

The Color Contour plot at the right in Figure 1 is similar to a contour map without labels on the contours. Each color traces out a constant elevation onthe surface. The elevation indicated by a given color on the 3D surface can be determined by the position of that color in the calibration scale at the bottomof the image.

(This program quantizes the range from the lowest to the highest elevation into 23 levels. Therefore, the accuracy of an elevation estimateis good to only about one 23rd of that total range. However, it would be an easy matter to increase the number of quantization levels used in thisprogram, thereby improving the accuracy of elevation estimates.)

For example, the blue contour that surrounds the central peak traces out the shape of an elevation that is about three levels up from the lowest elevation (as seen on the calibration scale) . The red contour that surrounds the central peak traces out the shape of an elevation that is about five levels down fromthe highest elevation. The aqua at the center of each of four minor peaks establishes their peak elevation to be about thirty-five percent of theelevation of the central peak (based on the position of aqua in the calibration scale) .