Bresenham circle drawing algorithm implementation in Java - java

I've implemented Bresenham Circle drawing algorithm in Java. But the output is not correctly drawn! I can't find where the problem is.
My code and output image is given below. Any help is appreciated.
public void display(GLAutoDrawable drawable) {
final GL2 gl = drawable.getGL().getGL2();
gl.glBegin (GL2.GL_POINTS);
double radius = 0.6;//sc.nextDouble();
double x =0.0;
double y = radius;
gl.glVertex2d(0.0,0.0);
gl.glVertex2d(x,y);
gl.glVertex2d(-x,y);
gl.glVertex2d(x,-y);
gl.glVertex2d(-x,-y);
gl.glVertex2d(y,x);
gl.glVertex2d(-y,x);
gl.glVertex2d(y,-x);
gl.glVertex2d(-y,-x);
double d = 5 - 4*radius;
while(x<y){
if(d<0){ //dE
x+=.01;
d+=(2*x + 3)*4;
}else{
x+=.01;
y-=.01;
d+=(2*x - 2*y +5)*4;
}
gl.glVertex2d(x,y);
gl.glVertex2d(-x,y);
gl.glVertex2d(x,-y);
gl.glVertex2d(-x,-y);
gl.glVertex2d(y,x);
gl.glVertex2d(-y,x);
gl.glVertex2d(y,-x);
gl.glVertex2d(-y,-x);
}
gl.glEnd();
}

Notice that the original Bresenham's circle algorithm works only with integers. Since your update is x_{n+1}=x_n+eps you can modify your y update to
y_{n+1}^2 = y_n^2 - 2*eps*n-eps*eps
The derivation is the same as the one given at the wiki page.
public void display(GLAutoDrawable drawable) {
final GL2 gl = drawable.getGL().getGL2();
gl.glBegin (GL2.GL_POINTS);
double radius = 0.6;//sc.nextDouble();
double x =0.0;
double y = radius;
gl.glVertex2d(0.0,0.0);
gl.glVertex2d(x,y);
gl.glVertex2d(-x,y);
gl.glVertex2d(x,-y);
gl.glVertex2d(-x,-y);
gl.glVertex2d(y,x);
gl.glVertex2d(-y,x);
gl.glVertex2d(y,-x);
gl.glVertex2d(-y,-x);
double eps = .01;
double eps2 = eps*eps;
while(x<y){
y = Math.sqrt(y*y-2*eps*x-eps2);
x+= eps;
gl.glVertex2d(x,y);
gl.glVertex2d(-x,y);
gl.glVertex2d(x,-y);
gl.glVertex2d(-x,-y);
gl.glVertex2d(y,x);
gl.glVertex2d(-y,x);
gl.glVertex2d(y,-x);
gl.glVertex2d(-y,-x);
}
gl.glEnd();
}
Result:
Be also careful of the aspect ratio. This algorithm works properly for ratio 1:1. If, however, your aspect ratio is a:b your equation for the circle would become x^2/a^2+y^2/b^2=r^2. You can change the update accordingly.

Here is example with drawing your custom circle by using Bresenham's algorithm.
Full code can be found in my repo : https://github.com/Maiakov/algorithms/tree/master/Task40
/**
* Write a routine to draw a circle (x ** 2 + y ** 2 = r ** 2) without making use of any floating point
* <p>
* computations at all.
*/
public class DrawCircleAlgorithm {
public static void drawCircle(int radius, int centerX, int centerY, Graphics g) {
int y = radius;
int x = 0;
int delta = calculateStartDelta(radius);
while (y >= x) {
drawPixelAndReflect(centerX, centerY, x, y, g);
if (delta < 0) {
delta = calculateDeltaForHorizontalPixel(delta, x);
} else {
delta = calculateDeltaForDiagonalPixel(delta, x, y);
y--;
}
x++;
}
}
private static int calculateStartDelta(int radius) {
return 3 - 2 * radius;
}
private static int calculateDeltaForHorizontalPixel(int oldDelta, int x) {
return oldDelta + 4 * x + 6;
}
private static int calculateDeltaForDiagonalPixel(int oldDelta, int x, int y) {
return oldDelta + 4 * (x - y) + 10;
}
private static void drawPixelAndReflect(int centerX, int centerY, int x, int y, Graphics g) {
g.drawLine(centerX + x, centerY + y, centerX + x, centerY + y);
g.drawLine(centerX + x, centerY - y, centerX + x, centerY - y);
g.drawLine(centerX - x, centerY + y, centerX - x, centerY + y);
g.drawLine(centerX - x, centerY - y, centerX - x, centerY - y);
g.drawLine(centerX - y, centerY + x, centerX - y, centerY + x);
g.drawLine(centerX - y, centerY - x, centerX - y, centerY - x);
g.drawLine(centerX + y, centerY + x, centerX + y, centerY + x);
g.drawLine(centerX + y, centerY - x, centerX + y, centerY - x);
}
}

Try this. I don't have java on the computer I am on right now, so let's see if it works. Make sure to work with integers since what you are normalizing are the block sizes!
Edit: Added integers.
public void display(GLAutoDrawable drawable) {
final GL2 gl = drawable.getGL().getGL2();
gl.glBegin (GL2.GL_POINTS);
double radius = 0.6;//sc.nextDouble();
double x =0.0;
double y = radius;
gl.glVertex2d(0.0,0.0);
gl.glVertex2d(x,y);
gl.glVertex2d(-x,y);
gl.glVertex2d(x,-y);
gl.glVertex2d(-x,-y);
gl.glVertex2d(y,x);
gl.glVertex2d(-y,x);
gl.glVertex2d(y,-x);
gl.glVertex2d(-y,-x);
double d = 3 - 2*radius;
while(x<y){
x++;
if(d<0){ //dE
d= d + 4x + 6;
}else{
y--;
d= d+ 4(x - y) + 10;
}
gl.glVertex2d(x,y);
gl.glVertex2d(-x,y);
gl.glVertex2d(x,-y);
gl.glVertex2d(-x,-y);
gl.glVertex2d(y,x);
gl.glVertex2d(-y,x);
gl.glVertex2d(y,-x);
gl.glVertex2d(-y,-x);
}
gl.glEnd();
}

Related

How to re format GetVertices to return (x1,y1,0), (x2,y2,0), (x3,y3,0);?

So, I have added a constructor that makes it possible to create triangles with the expression
new Triangle(x1, y1, x2, y2, x3, y3) where (x1,y1), (x2,y2), (x3,y3) are the three
vertices of the triangle. However, I need to make getVertices return (x1,y1,0), (x2,y2,0), (x3,y3,0); that is, the coordinates
given to the constructor, with coordinate z set to 0.
below is code for a triangular prism, im trying to shorten/reformat so it works for a triangle, with a different formula.
private List<Point> getVertices() {
List<Point> result = new ArrayList<>();
result.add(new Point(x1 + x, y1 + y, z - height / 2.0));
result.add(new Point(x2 + x, y2 + y, z - height / 2.0));
result.add(new Point(x3 + x, y3 + y, z - height / 2.0));
result.add(new Point(x1 + x, y1 + y, z + height / 2.0));
result.add(new Point(x2 + x, y2 + y, z + height / 2.0));
result.add(new Point(x3 + x, y3 + y, z + height / 2.0));
return result;
any help would be greatly appreciated
Point is a 2 dimensional coordinate of x and y. You would need to define your own class for a 3 dimensional x,y,z. Call the class Point3D
class Point3D {
int x, y, z;
public Point3D (int x, int y, int z) {
this.x = x;
this.y = y;
this.z = z;
}
public int getX() {
return x;
}
public int getY() {
return y;
}
public int getZ() {
return z;
}
}
That is not difficult but your List<Point3D> may not be compatible with something that expects List<Point> so you need to consider that.
You could also subclass the Point class. But there are quite a few methods you would need to add if you wanted the same functionality for a 3D point.

Coordinate values arent stored in processing

I really have no idea why the x and y values wont go to the drawLines function
float x, x1, x2;
float y, y1, y2;
float rad; //radius
int lines = 30; //number of lines
int colorNumber = 1;
void setup() {
background(#FFFFFF);
size (800, 600);
rad = 8;
}
void draw() {
}
This creates the three dots or vertices of the mathematical envelope
void mouseClicked() {
float x = mouseX;
float x1 = mouseX;
float x2 = mouseX;
float y = mouseY;
float y1 = mouseY;
float y2 = mouseY;
if (colorNumber == 1) {
fill(#9393ff);
ellipse(x, y, rad, rad);
} else if (colorNumber == 2) {
fill(#FF9393);
ellipse(x1, y1, rad, rad);
} else if (colorNumber == 3) {
fill(#93ff93);
ellipse(x2, y2, rad, rad);
}
}
This is supposed to draw the envelope using the coordinates of the vertices
void drawLines(int numLines) {
for (int i = 0; i < numLines; i = i + 1) {
float x = mouseX;
float x1 = mouseX;
float x2 = mouseX;
float y = mouseY;
float y1 = mouseY;
float y2 = mouseY;
float t = (float) i/(numLines-1);
float startX = x + t * (x1 - x);
float startY = y + t * (y1 - y);
float endX = x1 + t * (x2 - x1);
float endY = y1 + t * (y2 - y1);
line (startX, startY, endX, endY);
}
}
void mouseReleased() {
colorNumber++;
if (colorNumber == 4) {
colorNumber = 1;
}
println(colorNumber);
}
void keyPressed() {
if (keyPressed == true) {
background(#FFFFFF);
}
}
this last stuff just tells the code if you press a key, it will reset the backround
I understand your intention with using mouseX and mouseY to specify the coordinates of one of the 3 points of the envelope on click. The current issue is that all 3 points are being set to the same coordinate with each click. You need to introduce a variable to keep track of which coordinate to set on-click, such that only one pair is set. Then, only once all 3 coordinates are set, drawLines() can be called.
I propose the following:
Introduce 2 variables, one to keep track of which point is being modified; the other an array of PVectors (just to make it cleaner).
int index = 0;
PVector[] coords = new PVector[3];
Modify mouseClicked() to include the following:
void mouseClicked() {
ellipse(mouseX, mouseY, 8, 8);
coords[index] = new PVector(mouseX, mouseY);
index += 1;
if (index == 3) {
drawLines(lines);
}
index %= 3;
}
drawLines() becomes:
void drawLines(int numLines) {
for (int i = 0; i < numLines; i = i + 1) {
x = coords[0].x;
x1 = coords[1].x;
x2 = coords[2].x;
y = coords[0].y;
y1 = coords[1].y;
y2 = coords[2].y;
float t = (float) i / (numLines - 1);
float startX = x + t * (x1 - x);
float startY = y + t * (y1 - y);
float endX = x1 + t * (x2 - x1);
float endY = y1 + t * (y2 - y1);
line(startX, startY, endX, endY);
}
}
Finally, since your drawing on a black background, and the default stroke colour is black, use strokeColour() to change the colour of the lines so that you can see the envelope once its drawn.

Drawing a segment of a circle in Java?

I am making a brick breaker game in Java for fun. In this game the bat is a curved arc that goes around the circumference of a circle. I am struggling to make the bat behave properly.
I am drawing an arc that comes from 2 points on the circle:
public void update(){
if(dir == 1){
angle += 0.05;
}else if(dir == 0){
angle -= 0.05;
}
x0 = a + r * Math.cos(angle);
y0 = b + r * Math.sin(angle);
x1 = a + r * Math.cos(angle - 0.1);
y1 = b + r * Math.sin(angle - 0.1);
}
public void draw(Graphics2D g){
g.setColor(Color.black);
g.fillRect(0, 0, GamePanel.WIDTH, GamePanel.HEIGHT);
int tr = (int)Math.sqrt((x0-a)*(x0-a) + (y0-b)*(y0-b));
int x = (int) (a - tr);
int y = (int) (a - tr);
int width = 2*tr;
int height = 2*tr;
int startAngle = (int) (180/Math.PI*Math.atan2(y0-b, x0-a));
int endAngle = (int) (180/Math.PI*Math.atan2(y1-b, x1-a));
g.setColor(Color.white);
g.drawArc(x, y, width, height, startAngle, endAngle);
}
In theory this should work, the second points being generated from the angle going slightly further, but the length of the arc keeps varying in size...? That is where the problem lies.
This here statement breaks the pattern:
int y = (int) (a - tr);
It would make more sense to use
int y = (int) (b - tr);
And then there is the way g.drawArc is being called:
g.drawArc(x, y, width, height, startAngle, endAngle);
The last parameter is the angle of the arc, so I think you want
g.drawArc(x, y, width, height, startAngle, endAngle - startAngle );
possibly even
g.drawArc(x, y, width, height, startAngle, Math.abs(endAngle - startAngle) );

Basic bouncing ball program, ball won't bounce off wall

So I'm supposed to make a program where a ball bounces around a drawingpanel for 10 seconds. The ball has to bounce off the sides of the panel if it hits them. Right now when the ball hits the bottom panel instead of bouncing it appears in the middle of the screen and moves in the opposite direction until it hits the top and disappears.
I'm pretty sure the problem is in this part of my code...
(Earlier in the code I declared x to 1, y to 250, dx to 1, and dy to 1)
//Changes dirction
public static int newDirection1(int x, int dx, int size){
if (x < 0 || x > 500 || (x + size) < 0 || (x + size) > 500) {
dx *= -1;
return dx;
} else {
return dx;
}
}
//Changes direction
public static int newDirection2(int y, int dy, int size){
if (y < 0 || y > 500 || (y + size) < 0 || (y + size) > 500) {
dy *= -1;
return dy;
} else {
return dy;
}
}
//Moves ball one step
public static void move(Graphics g, Color color, int size, int x1, int y1, int x2, int y2){
g.setColor(Color.WHITE);
g.fillOval(x1, y1, size, size);
g.setColor(color);
g.fillOval(x2, y2, size, size);
}
//Pauses for 10ms
public static void sleep(int millis, DrawingPanel panel){
panel.sleep(millis);
}
public static void bounceLoop(DrawingPanel panel, Graphics g, Color color, int size, int x, int dx, int y, int dy, int millis){
int x1 = x + dx;
int x2 = x + dx;
int y1 = y + dy;
int y2 = y + dy;
for (int i = 0; i < 1000; i++) {
x1 = x + dx * i;
x2 = (x + dx * i) + dx;
y1 = y + dy * i;
y2 = (y + dy * i) + dy;
dx = newDirection1(x2, dx, size);
dy = newDirection2(y2, dy, size);
move(g, c, size, x1, y1, x2, y2);
sleep(millis, panel);
}
}
}
in the loop don't use:
x1 = x + dx * i
use
x1 = x1 + dx
(same for y)
because whenever dx is going to change, and multiply by -1, instead of continuing from where it was, and go to the other direction, it's going to continue from the other side of your panel, or a point that is really off.
Also a few things that could possibly fix the coding:
1- you don't need a dx parameter for your getNewDirection, you only need the coordinate.
2- the boundry conditions may give you errors, give it a small offset that can't be visible to the naked eye to avoid errors with creating objects outside the created panel or whatever you are using

Shoot to the mouse direction

The problem:
I've got this "Shot" class. In the code, the target variables are the mouseX and mouseY.
So when i click the mouse button, my player class will create a new shot object.
But the shooting is inaccurate.
How can i calculate the correct dx and dy?
If i add the dx and dy to the "bullet's" x and y, the bullet will move to the mouse's direction.This is what i want. The mouse position is stored in targetX and targetY, when the object is created. This is the point what the oval wants to reach.
Links:
The game (finished)
The code (from Shot.java):
public class Shot extends Entity {
private float targetX, targetY;
public Shot(World world, float x, float y, int width, int height, Color color, float targetX, float targetY) {
super(world, x, y, width, height, color);
this.targetX = targetX;
this.targetY = targetY;
}
#Override
public void render(GameContainer gc, Graphics g, Camera camera) {
g.setColor(color);
g.fillOval(x - camera.getX(), y - camera.getY(), width, height);
}
#Override
public void update(GameContainer gc, int delta) {
float dx = targetX - x;
float dy = targetY - y;
x += dx * delta * .001f;
y += dy * delta * .001f;
}
}
I tried this, but still not work:
#Override
public void update(GameContainer gc, int delta) {
float length = (float) Math.sqrt((targetX - x) * (targetX - x) + (targetY - y) * (targetY - y));
double dx = (targetX - x) / length * delta;
double dy = (targetY - y) / length * delta;
x += dx;
y += dy;
}
I did it! Here is my solution:
The problem was that, the target was the window's mouse position, and not the world's mouse position.
This is how i calculated the world's mouse positions:
float mouseWorldX = x + (mouseX - screen_width / 2); // x = player's x position
float mouseWorldY = y + (mouseY - screen_height / 2); // y = player's y position
This is code from my game at the moment is used to move a unit to the mouse when the right mouse button is pressed:
length = Math.sqrt((target_X - player_X)*(target_X - player_X) + (target_Y - player_Y)*(target_Y - player_Y)); //calculates the distance between the two points
speed_X = (target_X - player_X) /length * player_Speed;
speed_Y = (target_Y - player_Y) /length * player_Speed;
This will move an object to the target in a line at a set speed.
Edit: this is the actual code right from my game
if(input.isMouseButtonDown(Input.MOUSE_RIGHT_BUTTON))
{
length = (float) Math.sqrt((player_waypoint_X - player_X)*(player_waypoint_X - player_X) + (player_waypoint_Y - player_Y)*(player_waypoint_Y - player_Y));
velocityX = (float) (player_waypoint_X - player_X) /length * (float) PlayerStats.player.db_player_Speed;
velocityY = (float) (player_waypoint_Y - player_Y) /length * (float) PlayerStats.player.db_player_Speed;
player_waypoint_X = input.getMouseX() - 2;
player_waypoint_Y = input.getMouseY() - 2;
}
For testing purposes the velocity's are defined in the init method along with length. Every time the right mouse is pressed the waypoints's X and Y are changed to the mouse location.
I learned this from this question
velocity calculation algorithm.
in order to make the bullets not all change direction every shot, create an array list so that each bullet fired has its own x and y velocity

Categories

Resources