Range Selection

Setting LoBound and HiBound define the data range that gets colormapped. This can be useful for thresholding data to intensify the contrast in some regions while suppressing the display of other regions. The image below shows the GoldenGateDerivative.csv data in the HotCold colormap. This image was created using the Sobel edge detector in the horizontal and vertical directions and taking the root mean square of the results. As such, all the data displayed here is positive. The first image is the raw derivative data, the second has been thresholded to drop everything below 200.

Image of GoldenGateDerivative.csv in HotCold Colormap

Image of GoldenGateDerivative.csv in HotCold Colormap with LoBound = 200

Target Selection

The data reader utility comes with a few target-based colormaps that are good for highlighting regions of interest in the data. The black and white GoldenGate.jpg has values ranging from 0 to 255. Below the TargetRed colormap is used to highlight data in the 90-130 range. The histogram is shown below. Notice the spike is colored red while the rest of the data remains grayscale. The second image shows the pixel data with all the 90-130 pixels highlighted in red.

Histogram of GoldenGate.jpg with 90-130 range selected

GoldenGate.jpg displayed with Target colormap with 90-130 range selected

File Types

The colormap utility currently reads in three types of files: 1) Images (Bitmaps & JPEGs), 2) CSV (Comma Separated Values) Files, Colormap Images. This demonstration has shown the usage of images and CSV files. The final option, colormap images, is still in the experimental phase and is not completely stable yet.

The concept of reading colormapped images is that the user would select a bitmap that was saved using a colormap. The data reader would then parse the image into decimal values based on the color of each pixel. Depending on the colormap, the range of the data could encompass hundreds or thousands of separate values. The problem occurs with colormaps that have non-distinct color patterns. For instance, the GreenRedGreen colormap discussed earlier could not be used to extract data because for most colors two different values could be used.