Generate colors between red and green for a power meter?

Viewed 91613

I'm writing a Java game and I want to implement a power meter for how hard you are going to shoot something.

I need to write a function that takes a int between 0 - 100, and based on how high that number is, it will return a color between Green (0 on the power scale) and Red (100 on the power scale).

Similar to how volume controls work:
volume control

What operation do I need to do on the Red, Green, and Blue components of a color to generate the colors between Green and Red?

So, I could run say, getColor(80) and it will return an orangish color (its values in R, G, B) or getColor(10) which will return a more Green/Yellow RGB value.

I know I need to increase components of the R, G, B values for a new color, but I don't know specifically what goes up or down as the colors shift from Green-Red.


Progress:

I ended up using HSV/HSB color space because I liked the gradiant better (no dark browns in the middle).

The function I used was:

public Color getColor(double power)
{
    double H = power * 0.4; // Hue (note 0.4 = Green, see huge chart below)
    double S = 0.9; // Saturation
    double B = 0.9; // Brightness

    return Color.getHSBColor((float)H, (float)S, (float)B);
}

Where "power" is a number between 0.0 and 1.0. 0.0 will return a bright red, 1.0 will return a bright green.

Java Hue Chart:
Java Hue Chart

19 Answers

This should work - just linearly scale the red and green values. Assuming your max red/green/blue value is 255, and n is in range 0 .. 100

R = (255 * n) / 100
G = (255 * (100 - n)) / 100 
B = 0

(Amended for integer maths, tip of the hat to Ferrucio)

Another way to do would be to use a HSV colour model, and cycle the hue from 0 degrees (red) to 120 degrees (green) with whatever saturation and value suited you. This should give a more pleasing gradient.

Here's a demonstration of each technique - top gradient uses RGB, bottom uses HSV:

http://i38.tinypic.com/29o0q4k.jpg

Off the top of my head, here is the green-red hue transition in HSV space, translated to RGB:

blue = 0.0
if 0<=power<0.5:        #first, green stays at 100%, red raises to 100%
    green = 1.0
    red = 2 * power
if 0.5<=power<=1:       #then red stays at 100%, green decays
    red = 1.0
    green = 1.0 - 2 * (power-0.5)

The red, green, blue values in the above example are percentages, you'd probably want to multiply them by 255 to get the most used 0-255 range.

Linearly interpolating between green and red almost should work, except that in the middle there will be muddy brown color.

The most flexible and best looking solution is to have an image file somewhere that has the exact color ramp you want. And then lookup the pixel value in there. This has the flexibility that you can tweak the gradient to be just right.

If you still want to do it from code, then it's probably best to interpolate between green and yellow colors in the left side, and between yellow and red on the right side. In RGB space, green is (R=0, G=255, B=0), yellow is (R=255, G=255, B=0), red is (R=255, G=0, B=0) - this is assuming each color component goes from 0 to 255.

I made a small function wich gives you the rgb integer value for a percentage value:

private int getGreenToRedGradientByValue(int currentValue, int maxValue)
{
    int r = ( (255 * currentValue) / maxValue );
    int g = ( 255 * (maxValue-currentValue) ) / maxValue;
    int b = 0;
    return ((r&0x0ff)<<16)|((g&0x0ff)<<8)|(b&0x0ff);
}

So if your currentValue is 50 and your maxValue is 100, this function will return the color that you need, so if you loop this function with a percentage value, your color value will go from green to red. Sorry for the bad explanation

You need to linearly interpolate (LERP) the color components. Here's how it's done in general, given a start value v0, an end value v1 and the required ratio (a normalized float between 0.0 and 1.0):

v = v0 + ratio * (v1 - v0)

This gives v0 when ratio is 0.0, v1 when ratio is 1.0, and everything between in the other cases.

You can do this either on the RGB components, or using some other color scheme, like HSV or HLS. The latter two will be more visually pleasing, since they work on hue and brightness compoments that map better to our color perception.

Related