Related
I want to make a visual representation of the Pythagoras tree using Java, the code outputs a PNG fixed image.
I started by defining Vector class which starting from two vector components (x,y) can rotate the vector, scale it or add it to another vector.
public class Vector {
public double x;
public double y;
public Vector(double x, double y) {
this.x = x;
this.y = y;
}
public Vector rotated(double alpha) {
double x1 = Math.cos(alpha) * x - Math.sin(alpha) * y;
double y1 = Math.sin(alpha) * x + Math.cos(alpha) * y;
Vector vRotated = new Vector(x1, y1);
return vRotated;
}
public Vector scaled(double s) {
double x1 = x * s;
double y1 = y * s;
Vector vScaled = new Vector(x1, y1);
return vScaled;
}
public Vector added(Vector v) {
double x1 = this.x+v.x;
double y1 = this.y+v.y;
Vector vAdded = new Vector(x1,y1);
return vAdded;
}
}
I have also writen the method for creating the initial image and background and saving it to the desired path
public static void createPythagorasTreeImage(int startSize) throws IOException {
// Creation of the image object
int height = 5 * startSize;
int width = 8 * startSize;
BufferedImage image = new BufferedImage(width, height, BufferedImage.TYPE_INT_RGB);
// Create a Graphics2D object from the image and set a white background
Graphics2D g = image.createGraphics();
g.setColor(new Color(255, 255, 255));
g.fillRect(0, 0, width, height);
// Initial position and orientation of the first segment
Vector startPos = new Vector(width / 2, startSize);
Vector up = new Vector(0, 1);
// Start the recursion.
drawSegment(g, startPos, up, startSize, height);
// Save the image as PNG
String OS = System.getProperty("os.name").toLowerCase(); // different for win and unix
String filePath = System.getProperty("user.dir") + (OS.indexOf("win") >= 0 ? "\\" : "/") + "pythagorasTree.png";
System.out.println("Writing pythagoras-tree image to: " + filePath);
ImageIO.write(image, "png", new File(filePath));
}
I have read on wikipedia on how to the tree works, and want to now implement the algorithm.
What I need help with is implementing these two methods using Graphics2D (which I'm not very familiar with):
public static void drawRotatedRect(Graphics2D g, Vector pos, Vector up, int a, int height) {
}
This method should Draw a square using Graphics2D (maybe using g.fi llPolygon()?), at position pos, up the vector that indicates the rotation of the square by indicating which direction is up for the square, a is the side of the square and height is the height of the drawing space.
public static void drawSegment(Graphics2D g, Vector pos, Vector up, int a, int height) {
}
This method should draw the first square using the previous method, than compute the positions and rotations of the two new squares and draw them, repeat this recursively until a square has a very small side length (2px).
This is my understanding for the Pythagoras tree, I managed to write the majority of the code and it seems that the idea is correct, only if I get the two missing methods to work.
You can work with the Graphics2D context by drawing a Path2D with floating point (or double) precision. I reccoment this, since you will notice that using int precision might give you weird effects.
To draw a path, do:
Path2D.Double rectangle = new Path2D.Double();
rectangle.moveTo(0, 0);
// ... basically draw the four points of the rectangle here.
rectangle.closePath();
g.setColor(yourColorOfChoice);
g.fill(rectangle);
Notice that you need to draw the rectangular shapes manually, since they need ot be rotated, and Graphics2D does not do well with rotations. You could try using inherent rotations, but you will pixelate your context, and you won't like it.
I am very much looking forward to your results. Could you paste the final image into your question, once you are done :)?
i have an arraylist RecArray of objects with each object containing two int values, one for the width and for the height of a rectangle. Each rectangle's height and width are a multiple of ten. the rectangles have to be passed on to the surface as in the given order in RecArray from left to right and from top to bottom. my problem is i can not find the x,y coordinates of the next rectangle. what im trying to do is, starting at the coordinate (0,0) i generate the first rectangle, add it to an arraylist RecList. Then i set the x and y coordinates. x becomes x = x+RecArray.get(0).getLength1() + 1. if x is greater than the width of the jpanel surface then it becomes 0 and y becomes y = y + 10 . starting from the second object in the RecArray i try to generate rectangles with the given coordinates and width&height. Then i try to compare them with all the previous rectangles to see if there is any overlapping. if there is no overlapping, the rectangle will be drawn, if there is overlapping, the x coordinate of the rec becomes x = RecList.get(j).width+1 and if that exceeds the width x becomes 0 and y is y=y+10. Then i regenate the current rectangle with the new coordinates and compare with the other rectangles in RecList again till i find the right spot for the current rectangle.ive been dealing with that issue for the last 5 days and am really fed up now. i would greatly appreciate any tipps. and Please be patient with me. im still learning programming.
super.paintComponent(g);
Graphics2D g2 = (Graphics2D) g;
Rectangle rec = new Rectangle(x, y, RecArray.get(0).getWidth(),
RecArray.get(0).getHeight());
RecList.add(rec);
recPaint(g2,RecArray.get(0));
x = x + RecArray.get(0).getWidth() + 1;
int i;
for (i = 1; i < RecArray.size(); i++) {
if (x >= this.getArea().getWidth()) {
x = 0;
y = y + 10;
}
Rectangle rec1 = new Rectangle(x, y, RecArray.get(i)
.getWidth(), RecArray.get(i).getheight());
for (int j= 0; j < RecList.size(); j++) {
if (!recIntersect(rec1, RecList.get(j))) {
RecList.add(rec1);
recPaint(g2,RecArray.get(i));
break;
}
else {
x = RecList.get(j).width;
if (x >= this.getFlaeche().getLength1()) {
x = 0;
y = y + 10;
}
rec1 = new Rectangle(x, y,RecArray.get(i). .getWidth(),
RecArray.get(i).getHeight());
}
x = x + RecArray.get(i).getWidth();
}
//With this method using the given rec parameter a rectangle will be drawn on the g2 and filled in blue colour
private void recPaint (Graphics2D g2, RecType rec){
g2.setColor(Color.BLUE);
g2.fillRect(x, y, rec.getWidth(),
rec.getLength2());
g2.setColor(Color.BLACK);
g2.drawRect(x, y, rec.getHeight(),
rec.getLength2());
}
// returns true, if two rectangles overlap
private boolean recIntersect(Rectangle rec1, Rectangle rec2) {
if( rec1.intersects(rec2)){
return true;
}
return false;
}
Edit: apparently i haven't stated clearly what my problem is. my problem is, that the way i generate (x,y) coordinates of the rectangles is obviously wrong. the way my algorithm works doesnt get the results i want. i want my rectangles to be placed neatly next to/above/below each other WITHOUT overlapping, which is not the case.
Separate out your List of Rectangles. Calculate the X, Y coordinates once.
Since I didn't have your object class, I used the Dimension class, which holds a width and a length. I used the Rectangle class to hold the objects that will eventually be drawn in your Swing GUI.
Divide and conquer. Separate out your GUI model, view, and controller(s). This way, you can focus on one piece of the puzzle at a time.
Here are the results of my test code when I ran it with a drawing area of 500, 400.
java.awt.Rectangle[x=0,y=0,width=100,height=100]
java.awt.Rectangle[x=100,y=0,width=20,height=10]
java.awt.Rectangle[x=120,y=0,width=40,height=20]
java.awt.Rectangle[x=160,y=0,width=60,height=40]
java.awt.Rectangle[x=220,y=0,width=80,height=60]
java.awt.Rectangle[x=300,y=0,width=20,height=10]
java.awt.Rectangle[x=320,y=0,width=120,height=110]
Here are the results of my test code when I ran it with a drawing area of 200, 200.
java.awt.Rectangle[x=0,y=0,width=100,height=100]
java.awt.Rectangle[x=100,y=0,width=20,height=10]
java.awt.Rectangle[x=120,y=0,width=40,height=20]
java.awt.Rectangle[x=0,y=100,width=60,height=40]
java.awt.Rectangle[x=60,y=100,width=80,height=60]
java.awt.Rectangle[x=140,y=100,width=20,height=10]
And here's the code. I fit rectangles on the X axis until I can't fit another rectangle. Then I add the maximum height to Y, reset the X to zero, reset the maximum height and fit the next row of rectangles.
Create test applications like I did here and make sure that you can create the GUI model long before you create the GUI view and GUI controller.
package com.ggl.testing;
import java.awt.Dimension;
import java.awt.Rectangle;
import java.util.ArrayList;
import java.util.List;
public class CalculatingRectangles {
public static void main(String[] args) {
CalculatingRectangles calculatingRectangles = new CalculatingRectangles();
Dimension drawingArea = new Dimension(200, 200);
List<Dimension> dimensions = new ArrayList<>();
dimensions.add(new Dimension(100, 100));
dimensions.add(new Dimension(20, 10));
dimensions.add(new Dimension(40, 20));
dimensions.add(new Dimension(60, 40));
dimensions.add(new Dimension(80, 60));
dimensions.add(new Dimension(20, 10));
dimensions.add(new Dimension(120, 110));
List<Rectangle> rectangles = calculatingRectangles
.calculatingRectangles(drawingArea, dimensions);
System.out.println(displayRectangles(rectangles));
}
private static String displayRectangles(List<Rectangle> rectangles) {
StringBuilder builder = new StringBuilder();
for (Rectangle r : rectangles) {
builder.append(r);
builder.append(System.getProperty("line.separator"));
}
return builder.toString();
}
public List<Rectangle> calculatingRectangles(Dimension drawingArea,
List<Dimension> dimensions) {
int width = drawingArea.width;
int height = drawingArea.height;
int x = 0;
int y = 0;
int index = 0;
int maxHeight = 0;
boolean hasRoom = dimensions.size() > index;
List<Rectangle> rectangles = new ArrayList<>();
while (hasRoom) {
Dimension d = dimensions.get(index);
maxHeight = Math.max(maxHeight, d.height);
if ((x + d.width) <= width && (y + maxHeight) <= height) {
Rectangle r = new Rectangle(x, y, d.width, d.height);
x += d.width;
rectangles.add(r);
index++;
if (index >= dimensions.size()) {
hasRoom = false;
}
} else {
y += maxHeight;
if (y > height) {
hasRoom = false;
}
x = 0;
}
}
return rectangles;
}
}
I need to be able to rotate images individually(in java). The only thing I have found so far is g2d.drawImage(image, affinetransform, ImageObserver ). Unfortunately, I need to draw the image at a specific point, and there is no method with an argument that 1.rotates the image separately and 2. allows me to set the x and y. any help is appreciated
This is how you can do it. This code assumes the existance of a buffered image called 'image' (like your comment says)
// The required drawing location
int drawLocationX = 300;
int drawLocationY = 300;
// Rotation information
double rotationRequired = Math.toRadians (45);
double locationX = image.getWidth() / 2;
double locationY = image.getHeight() / 2;
AffineTransform tx = AffineTransform.getRotateInstance(rotationRequired, locationX, locationY);
AffineTransformOp op = new AffineTransformOp(tx, AffineTransformOp.TYPE_BILINEAR);
// Drawing the rotated image at the required drawing locations
g2d.drawImage(op.filter(image, null), drawLocationX, drawLocationY, null);
AffineTransform instances can be concatenated (added together). Therefore you can have a transform that combines 'shift to origin', 'rotate' and 'shift back to desired position'.
A simple way to do it without the use of such a complicated draw statement:
//Make a backup so that we can reset our graphics object after using it.
AffineTransform backup = g2d.getTransform();
//rx is the x coordinate for rotation, ry is the y coordinate for rotation, and angle
//is the angle to rotate the image. If you want to rotate around the center of an image,
//use the image's center x and y coordinates for rx and ry.
AffineTransform a = AffineTransform.getRotateInstance(angle, rx, ry);
//Set our Graphics2D object to the transform
g2d.setTransform(a);
//Draw our image like normal
g2d.drawImage(image, x, y, null);
//Reset our graphics object so we can draw with it again.
g2d.setTransform(backup);
I struggled a little with the existing answers because my image to be rotated is not always a square, furthermore the accepted answer has a comment asking "Any info on how to circumvent the cutoff problem" that is not answered.
So for those who had the issue of image being croped when rotated here is the code that worked for me :
public static BufferedImage rotate(BufferedImage bimg, Double angle) {
double sin = Math.abs(Math.sin(Math.toRadians(angle))),
cos = Math.abs(Math.cos(Math.toRadians(angle)));
int w = bimg.getWidth();
int h = bimg.getHeight();
int neww = (int) Math.floor(w*cos + h*sin),
newh = (int) Math.floor(h*cos + w*sin);
BufferedImage rotated = new BufferedImage(neww, newh, bimg.getType());
Graphics2D graphic = rotated.createGraphics();
graphic.translate((neww-w)/2, (newh-h)/2);
graphic.rotate(Math.toRadians(angle), w/2, h/2);
graphic.drawRenderedImage(bimg, null);
graphic.dispose();
return rotated;
}
public static BufferedImage rotateCw( BufferedImage img )
{
int width = img.getWidth();
int height = img.getHeight();
BufferedImage newImage = new BufferedImage( height, width, img.getType() );
for( int i=0 ; i < width ; i++ )
for( int j=0 ; j < height ; j++ )
newImage.setRGB( height-1-j, i, img.getRGB(i,j) );
return newImage;
}
from https://coderanch.com/t/485958/java/Rotating-buffered-image
Here is a solution for rotations of 90, 180 & 270 degrees.
For these cases, the AffineTransform can introduce some loss/interpolation.
This solution is lossless & can also handle (esoteric?) Colour Models with more than 8 Bits/Pixel which BufferedImage.getRGB(int x, int y) cannot.
The solution proceeds on a pixel-by-pixel basis, which has the advantage of being simple to code.
It is possible to read the original Image row-by-row to gain performance, but its more complex, so I've left that out.
import java.awt.image.BufferedImage;
import java.awt.image.WritableRaster;
/**
* N.B. this example uses the new switch/case/Arrow notation, which requires Java 14.
*/
public enum Rotation {
CLOCKWISE_90,
CLOCKWISE_180,
CLOCKWISE_270;
public BufferedImage rotate(final BufferedImage original) {
final int oW = original.getWidth();
final int oH = original.getHeight();
final BufferedImage rotated =
switch (this) {
case CLOCKWISE_180 -> new BufferedImage(oW, oH, original.getType());
default -> new BufferedImage(oH, oW, original.getType());
};
final WritableRaster rasterOriginal = original.copyData(null);
final WritableRaster rasterRotated = rotated .copyData(null);
/*
* The Data for 1 Pixel...
*/
final int[] onePixel = new int[original.getSampleModel().getNumBands()];
/*
* Copy the Pixels one-by-one into the result...
*/
for (int x = 0; x < oW; x++) {
for (int y = 0; y < oH; y++) {
; rasterOriginal.getPixel( x, y, onePixel);
switch (this) {
case CLOCKWISE_90 -> rasterRotated .setPixel(oH - 1 - y, x, onePixel);
case CLOCKWISE_270 -> rasterRotated .setPixel( y, oW - 1 - x, onePixel);
default -> rasterRotated .setPixel(oW - 1 - x, oH - 1 - y, onePixel);
};
}
}
rotated.setData(rasterRotated);
return rotated;
}
}
Sorry, but all the answers are difficult to understand for me as a beginner in graphics...
After some fiddling, this is working for me and it is easy to reason about.
#Override
public void draw(Graphics2D g) {
AffineTransform tr = new AffineTransform();
// X and Y are the coordinates of the image
tr.translate((int)getX(), (int)getY());
tr.rotate(
Math.toRadians(this.rotationAngle),
img.getWidth() / 2,
img.getHeight() / 2
);
// img is a BufferedImage instance
g.drawImage(img, tr, null);
}
I suppose that if you want to rotate a rectangular image this method wont work and will cut the image, but I thing you should create square png images and rotate that.
I think the sturdiest and easiest approach is to not only rotate the image, but also the coordinates you're working with. This code will turn a bufferedImage around the topleft corner and draw it accordingly without cropping, and the returned image will be drawn in the right place. If you know what a vector and matrix are and look at the wikipedia article for rotation matrix you will understand this code easily. I also added a little hand drawing to it. In the 2nd part of the algorithm we determine the position and dimension of the bigger rectangle, which contains our rotated bufferedImage. The first 4 points we define are technically not used, but they still help you understand where the points are initially.
public void drawRotated(final Graphics g, final BufferedImage bufferedImage, final int x, final int y, final int width, final int height, final double angle) {
final Rectangle collision = new Rectangle(x, y, width, height);
final BufferedImage resize = resize(bufferedImage, collision.width, collision.height);
final BufferedImage rotate = rotate(resize, angle, collision);
g.drawImage(rotate, collision.x, collision.y, collision.width, collision.height, null);
}
public static BufferedImage resize(final BufferedImage bufferedImage, final int newWidth, final int newHeight) {
final BufferedImage resized = new BufferedImage(newWidth, newHeight, bufferedImage.getType());
final Graphics g = resized.createGraphics();
g.drawImage(bufferedImage, 0, 0, newWidth, newHeight, null);
return resized;
}
public static BufferedImage rotate(final BufferedImage bufferedImage, final double angle, final Rectangle collision) {
final double sin = Math.sin(Math.toRadians(angle));
final double cos = Math.cos(Math.toRadians(angle));
final int x1 = collision.x;
final int y1 = collision.y;
final int x2 = collision.x+collision.width;
final int y2 = collision.y;
final int x3 = collision.x;
final int y3 = collision.y+collision.height;
final int x4 = collision.x+collision.width;
final int y4 = collision.y+collision.height;
//turn all 4 points around the top left point
final int newx1 = collision.x;
final int newy1 = collision.y;
//the y component is 0
final int newx2 = (int) (collision.x+collision.width*cos);
final int newy2 = (int) (collision.y+collision.width*sin);
//the x component is 0
final int newx3 = (int) (collision.x-collision.height*sin);
final int newy3 = (int) (collision.y+collision.height*cos);
final int newx4 = (int) (collision.x+collision.width*cos-collision.height*sin);
final int newy4 = (int) (collision.y+collision.width*sin+collision.height*cos);
//determine the new position of our bigger rectangle containing our image
collision.x = Math.min(Math.min(newx1, newx2), Math.min(newx3, newx4));
collision.y = Math.min(Math.min(newy1, newy2), Math.min(newy3, newy4));
//determine the new dimensions of our bigger rectangle containing our image
collision.width = Math.max(Math.max(newx1, newx2), Math.max(newx3, newx4))-collision.x;
collision.height = Math.max(Math.max(newy1, newy2), Math.max(newy3, newy4))-collision.y;
final BufferedImage rotated = new BufferedImage(collision.width, collision.height, bufferedImage.getType());
final Graphics2D g2d = rotated.createGraphics();
g2d.translate(newx1- collision.x, newy1- collision.y);
g2d.rotate(Math.toRadians(angle), 0, 0);
g2d.drawRenderedImage(bufferedImage, null);
g2d.dispose();
return rotated;
}
I have a simple game animation made in java. It is of three planets rotating around an axis. Each planet is an instance of the class Planet and they have an update method which, every time it is run, the orbit's rotation angle increases and the position is updated acording to the angle and a few predetermined variables like distance from the "sun". From here, you can determine the position of the planet with simple trigonometry. In this case:
Sin(angle) = op/hyp = y/distance
therefore
Sin(angle)*hyp = op
Cos(angle) = ady/hyp = x/distance
therefore
Cos(angle)*hyp = ady
where the hypothenuse is the distance to the sun and the adyacent and oposite sides are the x and y values respectively. I figured this would work, until I tried it out. It gave me an eliptical rotation. Here is the code that updates the planet's rotation (orbit center is the sun's center position):
position.x = ((Math.cos(orbitAngle) * orbitDistance) + orbitCenter.x);
position.y = ((Math.sin(orbitAngle) * orbitDistance) + orbitCenter.y);
What could be wrong?
EDIT:
I realized this problem by placing an object with its center in the position specified by orbit center
Here is the full code of the planet class:
public class Planet
{
protected Image image;
protected Vector2 position;
protected final Vector2 orbitCenter;
protected float rotation;
protected Vector2 imageSize;
protected final float rotationSpeed;
protected final float orbitDistance;
protected float orbitAngle;
protected final float orbitAngleSpeed;
public Planet(Image image, float orbitDistance, float rotationSpeed, Vector2 orbitCenter, float orbitAngleSpeed)
{
this.image = image;
this.position = new Vector2(orbitCenter.x, orbitCenter.y - orbitDistance);
this.orbitCenter = orbitCenter;
this.rotation = 0;
this.imageSize = new Vector2(image.getWidth(null), image.getHeight(null));
this.rotationSpeed = rotationSpeed;
this.orbitDistance = orbitDistance;
this.isMouseOver = false;
this.isPressed = false;
this.orbitAngle = 0;
this.orbitAngleSpeed = orbitAngleSpeed;
}
public void Update()
{
orbitAngle += orbitAngleSpeed;
if(orbitAngle > Math.PI * 2)
orbitAngle %= Math.PI * 2;
position.x = ((Math.cos(orbitAngle) * orbitDistance) + orbitCenter.x);
position.y = ((Math.sin(orbitAngle) * orbitDistance) + orbitCenter.y);
}
public void Draw(Graphics2D g)
{
g.rotate(rotation, position.x + imageSize.x / 2, position.y + imageSize.y / 2);
g.drawImage(image, (int)position.x, (int)position.y, null);
g.rotate(-rotation, position.x + imageSize.x / 2, position.y + imageSize.y / 2);
}
}
Here is the class that tests the planet class. You can download the jar it needs to work from here: foxtailgames.net/AppletSource.jar. Here is the tester class (you will probably have to import a few things though if you do it in eclipse or netbeans it will give you the imports):
public class PlanetTest extends AppletCore
{
public void resizeScreen() {resize(800, 800);}
Image center;
Planet p;
public void LoadContent()
{
p = new Planet(loadImage("images/GameMenuCircles/Planet1.png"), 100f, 0.02f, new Vector2(400, 400), 0.005f);
center = loadImage("images/GameMenuCircles/Center.png");
}
public void Update(GameTime gameTime)
{
p.Update();
}
public void Draw(Graphics2D g, GameTime gameTime)
{
g.drawImage(center, 400 - center.getWidth(null)/2, 400 - center.getWidth(null)/2, null);
p.Draw(g);
g.setColor(Color.green);
g.drawLine(400, 400, 500, 400);
g.drawLine(400, 400, 400, 500);
g.drawLine(400, 400, 300, 400);
g.drawLine(400, 400, 400, 300);
g.setColor(Color.white);
}
}
Your rotation is set to 0 in the above so i assume you are not rotating the picture at the moment. What i think is happening is the orbit circle you are producing is fine, but the location you are drawing the planet is off.
Below is an image of how Swing would draw the circle, so the overlap you experience is because of this.
You need to adjust the position you draw the circle by how half the width so it sits over the center of the orbit.
EDIT: You've alter some code but what you need to change is the draw method of he planet:
public void Draw(Graphics2D g) {
g.rotate(rotation, position.x + imageSize.x / 2, position.y + imageSize.y / 2);
g.drawImage(image, (int)position.x, (int)position.y, null); //here
g.rotate(-rotation, position.x + imageSize.x / 2, position.y + imageSize.y / 2);
}
This line needs to be:
g.drawImage(image, (int)position.x - imageSize.width, (int)position.y - imageSizee.height, null); //here
You might compare your result to this AnimationTest that uses the same parametric equation of a circle. Because the orbital radius is a function of the enclosing panel's dimensions, the orbit is circular only when w equals h. Resize the frame, or set HIGH = WIDE, to see the effect.
I am currently attempting to draw images on the screen at a regular rate like in a video game.
Unfortunately, because of the rate at which the image is moving, some frames are identical because the image has not yet moved a full pixel.
Is there a way to provide float values to Graphics2D for on-screen position to draw the image, rather than int values?
Initially here is what I had done:
BufferedImage srcImage = sprite.getImage ( );
Position imagePosition = ... ; //Defined elsewhere
g.drawImage ( srcImage, (int) imagePosition.getX(), (int) imagePosition.getY() );
This of course thresholds, so the picture doesn't move between pixels, but skips from one to the next.
The next method was to set the paint color to a texture instead and draw at a specified position. Unfortunately, this produced incorrect results that showed tiling rather than correct antialiasing.
g.setRenderingHint ( RenderingHints.KEY_ANTIALIASING, RenderingHints.VALUE_ANTIALIAS_ON );
BufferedImage srcImage = sprite.getImage ( );
g.setPaint ( new TexturePaint ( srcImage, new Rectangle2D.Float ( 0, 0, srcImage.getWidth ( ), srcImage.getHeight ( ) ) ) );
AffineTransform xform = new AffineTransform ( );
xform.setToIdentity ( );
xform.translate ( onScreenPos.getX ( ), onScreenPos.getY ( ) );
g.transform ( xform );
g.fillRect(0, 0, srcImage.getWidth(), srcImage.getHeight());
What should I do to achieve the desired effect of subpixel rendering of an Image in Java?
You can use a BufferedImage and AffineTransform, draw to the buffered image, then draw the buffered image to the component in the paint event.
/* overrides the paint method */
#Override
public void paint(Graphics g) {
/* clear scene buffer */
g2d.clearRect(0, 0, (int)width, (int)height);
/* draw ball image to the memory image with transformed x/y double values */
AffineTransform t = new AffineTransform();
t.translate(ball.x, ball.y); // x/y set here, ball.x/y = double, ie: 10.33
t.scale(1, 1); // scale = 1
g2d.drawImage(image, t, null);
// draw the scene (double percision image) to the ui component
g.drawImage(scene, 0, 0, this);
}
Check my full example here: http://pastebin.com/hSAkYWqM
You can composite the image yourself using sub-pixel accuracy, but it's more work on your part. Simple bilinear interpolation should work well enough for a game. Below is psuedo-C++ code for doing it.
Normally, to draw a sprite at location (a,b), you'd do something like this:
for (x = a; x < a + sprite.width; x++)
{
for (y = b; y < b + sprite.height; y++)
{
*dstPixel = alphaBlend (*dstPixel, *spritePixel);
dstPixel++;
spritePixel++;
}
dstPixel += destLineDiff; // Move to start of next destination line
spritePixel += spriteLineDiff; // Move to start of next sprite line
}
To do sub-pixel rendering, you do the same loop, but account for the sub-pixel offset like so:
float xOffset = a - floor (a);
float yOffset = b - floor (b);
for (x = floor(a), spriteX = 0; x < floor(a) + sprite.width + 1; x++, spriteX++)
{
for (y = floor(b), spriteY = 0; y < floor (b) + sprite.height + 1; y++, spriteY++)
{
spriteInterp = bilinearInterp (sprite, spriteX + xOffset, spriteY + yOffset);
*dstPixel = alphaBlend (*dstPixel, spriteInterp);
dstPixel++;
spritePixel++;
}
dstPixel += destLineDiff; // Move to start of next destination line
spritePixel += spriteLineDiff; // Move to start of next sprite line
}
The bilinearInterp() function would look something like this:
Pixel bilinearInterp (Sprite* sprite, float x, float y)
{
// Interpolate the upper row of pixels
Pixel* topPtr = sprite->dataPtr + ((floor (y) + 1) * sprite->rowBytes) + floor(x) * sizeof (Pixel);
Pixel* bottomPtr = sprite->dataPtr + (floor (y) * sprite->rowBytes) + floor (x) * sizeof (Pixel);
float xOffset = x - floor (x);
float yOffset = y - floor (y);
Pixel top = *topPtr + ((*(topPtr + 1) - *topPtr) * xOffset;
Pixel bottom = *bottomPtr + ((*(bottomPtr + 1) - *bottomPtr) * xOffset;
return bottom + (top - bottom) * yOffset;
}
This should use no additional memory, but will take additional time to render.
I successfully solved my problem after doing something like lawrencealan proposed.
Originally, I had the following code, where g is transformed to a 16:9 coordinate system before the method is called:
private void drawStar(Graphics2D g, Star s) {
double radius = s.getRadius();
double x = s.getX() - radius;
double y = s.getY() - radius;
double width = radius*2;
double height = radius*2;
try {
BufferedImage image = ImageIO.read(this.getClass().getResource("/images/star.png"));
g.drawImage(image, (int)x, (int)y, (int)width, (int)height, this);
} catch (IOException ex) {
Logger.getLogger(View.class.getName()).log(Level.SEVERE, null, ex);
}
}
However, as noted by the questioner Kaushik Shankar, turning the double positions into integers makes the image "jump" around, and turning the double dimensions into integers makes it scale "jumpy" (why the hell does g.drawImage not accept doubles?!). What I found working for me was the following:
private void drawStar(Graphics2D g, Star s) {
AffineTransform originalTransform = g.getTransform();
double radius = s.getRadius();
double x = s.getX() - radius;
double y = s.getY() - radius;
double width = radius*2;
double height = radius*2;
try {
BufferedImage image = ImageIO.read(this.getClass().getResource("/images/star.png"));
g.translate(x, y);
g.scale(width/image.getWidth(), height/image.getHeight());
g.drawImage(image, 0, 0, this);
} catch (IOException ex) {
Logger.getLogger(View.class.getName()).log(Level.SEVERE, null, ex);
}
g.setTransform(originalTransform);
}
Seems like a stupid way of doing it though.
Change the resolution of your image accordingly, there's no such thing as a bitmap with sub-pixel coordinates, so basically what you can do is create an in memory image larger than what you want rendered to the screen, but allows you "sub-pixel" accuracy.
When you draw to the larger image in memory, you copy and resample that into the smaller render visible to the end user.
For example: a 100x100 image and it's 50x50 resized / resampled counterpart:
See: http://en.wikipedia.org/wiki/Resampling_%28bitmap%29