Page 231 - The Definitive Guide to Building Java Robots

Preston_5564C06.fm Page 212 Friday, September 23, 2005 5:13 AM

212 CHAPTER 6 ■ VISION

// sample program
public static void main(String[] args) {
try {
// load image
ColorGramCalibration cg2 = new ColorGramCalibration(
"sample_images//coke.jpg", true);
// optimize it
cg2.optimize();
// print colorgram for cut-paste if needed
System.out.println(cg2.getBestColorGram().toString());
} catch (Exception e) {
e.printStackTrace();
System.exit(1);
}
}

}

Figure 6-20. All ColorGrams after optimization (Coke, 7-Up, and Pepsi)

Section Summary

In this section, I showed you a few ways to use color to identify objects. For my goal of getting
a beverage from the refrigerator, this was all I needed. The six methods and two classes created
in this section were
• ImageProcessor.thresholdColor(): A method to get the color thresholds of certain
minimum and maximum values for a particular color band: RED, GREEN, or BLUE
• ImageProcessor.filterColor(): A method to filter colors of a particular band

• ImageProcessor.getMean(): A method to compute some image statistics and get the
average colors of each band

• ImageProcessor.getHistogram(): A method that offers an alternative method of getting
the same statistics using Java Advanced Imaging and the histogram operation
• ImageProcessor.colorRatio(): A method that processes an image by colors’ ratios
between one another
• ImageProcessor.colorRatioCount(): A method that counts the colors in a ratio