I am making a tiled based game in java and I want to make a light map.
I am having some issues. I have the lightmap array that has lights placed on it that affect the array. Lights emit in a circle shape. It seems ok so far but its not exactly what I wanted.
Here is my code so far:
for(float i = 0; i < strength + 1; i++){
for(double u = 0.0f; u < 360; u += 0.5){
double angle = u * Math.PI / 180;
int x2 = (int)(x + i * Math.cos(angle));
int y2 = (int)(y + i * Math.sin(angle));
if(map[y2][x2] > 1 - 1 / i)
map[y2][x2] = 1 - 1 / i;
}
}
Result:
As you can see in the result, it seems as though the light is expanding too much on the bottom left side (red x's). How do I fix this?
Background info:
Strength:
The radius of how far the light reaches. This also
determines how bright the light will be at each tile of the array.
The Array "map" is a 2D float array. The engine I am using uses float
values for the alpha channel. The range is 0 (completely transparent)
to 1 (completely opaque).
Solution (Thanks to Gene):
for(int x2 = -strength; x2 <= strength; x2++){
for (int y2 = -strength; y2 <= strength; y2++) {
double r = Math.sqrt(x2 * x2 + y2 * y2);
double inv_rad = r <= strength + 1 ? 1 / r : 0;
if(map[y + y2][x + x2] > 1 - (float) inv_rad)
map[y + y2][x + x2] = 1 - (float) inv_rad;
}
}
Your algorithm suffers from integer truncation of the map indicies. Try it the other away around. Compute the distance from each pixel in a square surrounding the center to the center. From this distance calculate what the intensity ought to be. It will be something like this:
for (x = -R; x <= R; x++)
for (y = -R; y <= R; y++) {
double r = Math.sqrt(x * x + y * y);
double inv_rad = r <= R ? 1 / r : 0; // truncate outside radius R
map[yc + y][xc + x] = 1 - inv_rad;
}
Here xc and yc are the integer center coordinates. R is the half-size of the box around the center.
when i try to add this to my project i only get o.o back
the values i entered where 500, 500,50
private float map[][] = new float[1000][1000];
public void test(int x, int y, float strength){
public void addLight(int x,int y,int strength ){
for(int x2 = -strength; x2 <= strength; x2++){
for (int y2 = -strength; y2 <= strength; y2++) {
double r = Math.sqrt(x2 * x2 + y2 * y2);
double inv_rad = r <= strength + 1 ? 1 / r : 0;
if(map[y + y2][x + x2] > 1 - (float) inv_rad)
map[y + y2][x + x2] = 1 - (float) inv_rad;
System.out.println(map[y + y2][x + x2]);
}
}
}
Related
I am trying to hollow out a sphere, I already achieved this however it is incredibly slow (A few seconds for a sphere with a radius of 5, a few minutes for a sphere with a radius of 100)
This is my code:
BlockPos pos = player.getBlockPos();
int startX = pos.getX();
int startY = pos.getY();
int startZ = pos.getZ();
int squaredRadius = radius * radius;
int x1;
int y1;
int z1;
int flags = 2 | 8 | 16 | 32;
BlockPos.Mutable blockPos = new BlockPos.Mutable(0, 0, 0);
BlockState state = Blocks.AIR.getDefaultState();
for (int x = startX - radius; x < startX + radius; x++) {
for (int y = Math.max(0, startY - radius), maxY2 = Math.min(255, startY + radius); y < maxY2; y++) {
for (int z = startZ - radius; z < startZ + radius; z++) {
x1 = x - startX;
y1 = y - startY;
z1 = z - startZ;
if (x1 * x1 + y1 * y1 + z1 * z1 <= squaredRadius)
world.setBlockState(blockPos.set(x, y, z), state, flags);
}
}
}
How could I speed this up? (Or is it already as fast as it can be?)
EDIT: For a 100 radius it takes roughly 8 minutes on my machine
Update: I managed to fix it! What I did was make a mixin into World#setBlockState(BlockPos, BlockState, int) and check if the flags were some number I set (999 in my case), and if it where I canceled the method execution right after the block state was set, saving huge amounts of time when placing a lot of blocks. However, it did not send the chunks to the player so what you have to do is simply send all the chunks you changed to the player, I did this by storing all the chunks I changed and for each player in the world I sent a ChunkDataS2CPacket with the new chunk to the player, which fixed my problem entirely, it works like a charm :D
EDIT: It also does not do block updates/lighting updates
I have to points in 3D space that I want to draw a cylinder between. I currently have this code:
applet.pushMatrix();
applet.stroke(0);
applet.fill(0);
applet.line(this.start.x, this.start.y, this.start.z, this.end.x, this.end.y, this.end.z); //debug, shows where the cylinder should be
applet.translate(this.start.x, this.start.y);
applet.beginShape(PConstants.TRIANGLE_STRIP);
float xdif = this.end.x - this.start.x;
float ydif = this.end.y - this.start.y;
float zdif = this.end.z - this.start.z;
float rx = (float)Math.atan(ydif / xdif);
float ry = (float)Math.atan(zdif / xdif);
float rz = (float)Math.atan(zdif == 0 ? 0 : ydif / zdif);
applet.rotateZ((float) (rx - Math.PI / 2));
applet.rotateY((float) (ry - Math.PI / 2));
applet.rotateX((float) (rz - Math.PI / 2));
float r = 20;
float len = HelperFunctions.distance(this.start, this.end) / 2;
boolean w = false;
for (int i = 0; i < 2; i += 1) {
for (float j = 0; j < Math.PI * 2 + 0.2; j += 0.4) {
w = !w;
float z = (i + (w ? 0 : 1)) * len;
float x = (float)(r * Math.cos(j));
float y = (float)(r * Math.sin(j));
applet.vertex(y, x, z);
}
}
applet.endShape();
applet.popMatrix();
The cylinder does draw correctly, however, it doesn't rotate correctly. I'm trying to use trig to determine the rotation angles, however, I'm not sure I've done it correctly. How would I get the correct angles to rotate around such that it is drawn from Vector start to Vector end?
Vector is a custom class with just a float x, y, z, and HelperFunctions.distance takes two Vectors and calculates the distance (pythagoras).
Thanks in advance.
I have a 3D model and I need to rotate its vertices around the Y axis (The axis going straight up in my case). For example lets say i have the vert
(3,2,3)(x,y,z) and when i rotate around the Y axis only the x and z's will change. how could I implement this in java using degrees? Thanks in advance!
(FYI) this is for rotating the points on my hitbox. Each "box" is just a triangle but wrapped in a cube so i can just check if a point is in the cube. This is done per triangle per model. This works perfectly because im able to walk through meshes with holes in them and everything. However, if any rotation is applied weird things start to happen.
Edit: here is my code using Andys method
public static boolean checkPointCollision(Vector3f pos){
boolean hit=false;
float px=Math.round(pos.x);
float py=Math.round(pos.y);
float pz=Math.round(pos.z);
px=pos.x;
py=pos.y;
pz=pos.z;
long startTime=System.currentTimeMillis();
float xmin,ymin,zmin,xmax,ymax,zmax,scale,rot;
//Cube Collisions
for (Entity entity : entities) {
int colID=entity.getCollisionIndex();
boolean entHasHitbox = entity.hasHitbox();
if(colID!=-1 && hit==false && entHasHitbox){
//Gets the entitys variables
scale = entity.getScale();
rot = entity.getRotY();
//Converts to radians
rot = (float) Math.toRadians(rot);
xmin = 0;
ymin = 0;
zmin = 0;
xmax = 0;
ymax = 0;
zmax = 0;
switch(entity.getCollisionType()){
case 1:
if(entHasHitbox){
//Gets the entities hitbox
List<Vector3f> hitboxMins = entity.getHitboxMin();
List<Vector3f> hitboxMaxs = entity.getHitboxMax();
for (int i = 0; i < hitboxMins.size(); i++) {
//Gets the entities hitbox points
Vector3f min = hitboxMins.get(i);
Vector3f max = hitboxMaxs.get(i);
//Sets all local position vars to the hitboxes mins and maxes
xmin = min.x;
ymin = min.y;
zmin = min.z;
xmax = max.x;
ymax = max.y;
zmax = max.z;
//Applies the models scale
xmin *=scale;
ymin *=scale;
zmin *=scale;
xmax *=scale;
ymax *=scale;
zmax *=scale;
//Rotates points
float nxmin = (float) (Math.cos(rot) * xmin - Math.sin(rot) * zmin);
float nzmin = (float) (Math.sin(rot) * xmin + Math.cos(rot) * zmin);
float nxmax = (float) (Math.cos(rot) * xmax - Math.sin(rot) * zmax);
float nzmax = (float) (Math.sin(rot) * xmax + Math.cos(rot) * zmax);
//Sets old points to new ones
xmin = nxmin;
zmin = nzmin;
xmax = nxmax;
zmax = nzmax;
//Increase local points to the entitys world position
xmin += entity.getPosition().x;
xmax += entity.getPosition().x;
ymin += entity.getPosition().y;
ymax += entity.getPosition().y;
zmin += entity.getPosition().z;
zmax += entity.getPosition().z;
//Debug
if(entities.get(17)==entity){//entities.get(17).increaseRotation(0, 10, 0);
System.out.println(xmin+","+ymin+","+zmin);
}
//Check if point is in the hitbox
if(px>=xmin && px<=xmax
&& py>=ymin && py<=ymax
&& pz>=zmin && pz<=zmax)
{
hit=true;
//Ends to loop
i=hitboxMins.size();
}
}
}
break;
}
}
}
long endTime = System.currentTimeMillis()-startTime;
if(endTime>10){
System.out.println("Delay in Point Collision");
}
return hit;
}
Multiply your points by the following matrix:
[ c 0 -s ]
[ 0 1 0 ]
[ s 0 c ]
i.e.
[newx] [ c 0 -s ] [x]
[newy] = [ 0 1 0 ] [y]
[newz] [ s 0 c ] [z]
where (x, y, z) are your original coordinates, (newx, newy, newz) are your rotated coordinates, and c = cos(angle) and s = sin(angle). Note that Java's trig functions take their parameters as radians, so you need to convert the angle in degrees appropriately.
If you've not used matrices before, this is equivalent to the following three expressions:
newx = c * x - s * z
newy = y
newz = s * x + c * z
Let's first start off with what I am trying to do. I would like to be able to take PNG file with a transparent background and find anywhere from 90 to 360 points along the edge of the subject of the image. Here is a rough example of what I mean. Given this image of Mario and Yoshi:
I want to make a circle that is centered at the center of the image with a diameter slightly larger than the largest side of the image to serve as a reference. Then, I want to go around the circle at set intervals, and trace a line towards the center until it hits a non-transparent pixel. Here is what that would look like:
I have attempted to implement this a few different times, all of which failed, and I was hoping to get some guidance or insight as to what I am doing wrong. Here is an image of the math I am using behind the code (sorry if the quality is not great, I don't have a scanner):
The Line 1 is either the top, bottom, left or right edge of the image, and Line 2 goes through the center of the circle at the given angle. The point where lines 1 and 2 intersect should be on the edge of the image, and is where we should start looking for the edge of the image's subject.
Here is the code that I came up with from this idea. I did it in Java because BufferedImage is really easy to use, but I am going to translate this over to C# (XNA) for the final product.
public class Mesh {
private int angleA, angleB, angleC, angleD;
private BufferedImage image;
private Point center;
public ArrayList<Point> points = new ArrayList<>();
public Mesh(BufferedImage image) {
center = new Point(image.getWidth() / 2, image.getHeight() / 2);
angleA = (int) (Math.atan(center.y / center.x) * (180 / Math.PI));
angleB = 180 - angleA;
angleC = 180 + angleA;
angleD = 360 - angleA;
this.image = image;
for(int angle = 0; angle <= 360; angle+=4){
Point point = getNext(angle);
if(point != null) points.add(point);
}
}
private Point getNext(int angle) {
double radians = angle * Math.PI / 180;
double xStep = Math.cos(radians);
double yStep = Math.sin(radians);
int addX = angle >= 90 && angle <= 270 ? 1 : -1;
int addY = angle >= 0 && angle <= 180 ? 1 : -1;
double x, y;
if (xStep != 0) {
double slope = yStep / xStep;
double intercept = center.y - (slope * center.x);
if (angle >= angleA && angle <= angleB) {
y = 0;
x = -intercept / slope;
} else if (angle > angleB && angle < angleC) {
x = 0;
y = intercept;
} else if (angle >= angleC && angle <= angleD) {
y = image.getHeight() - 1;
x = (y - intercept) / slope;
} else {
x = image.getWidth() - 1;
y = slope * x + intercept;
}
} else {
x = center.x;
y = angle <= angleB ? 0 : image.getHeight();
}
if (x < 0) x = 0;
if (x > image.getWidth() - 1) x = image.getWidth() - 1;
if (y < 0) y = 0;
if (y > image.getHeight() - 1) y = image.getHeight() - 1;
double distance = Math.sqrt(Math.pow(x - center.x, 2) + Math.pow(y - center.y, 2));
double stepSize = Math.sqrt(Math.pow(xStep, 2) + Math.pow(yStep, 2));
int totalSteps = (int) Math.floor(distance / stepSize);
for (int step = 0; step < totalSteps; step++) {
int xVal = (int) x;
int yVal = (int) y;
if(xVal < 0) xVal = 0;
if(xVal > image.getWidth() -1) xVal = image.getWidth() -1;
if(yVal < 0) yVal = 0;
if(yVal > image.getHeight()-1) yVal = image.getHeight() -1;
int pixel = image.getRGB(xVal, yVal);
if ((pixel >> 24) == 0x00) {
x += (Math.abs(xStep) * addX);
y += (Math.abs(yStep) * addY);
} else {
return new Point(xVal, yVal);
}
}
return null;
}
}
The algorithm should be returning all positive points that are all ordered in counterclockwise rotation (and non-overlapping) but I have failed to get the desired output (this being my most recent attempt) so just to restate the question, is there a formalized way of doing this, or can someone find the mistake I made in my logic. For visual reference, the Mario and Yoshi Traced image is sort of what the final output should look like, but with many more points (which would give more detail to the mesh).
I am trying to find image in an image. I do this for desktop automation. At this moment, I'm trying to be fast, not precise. As such, I have decided to match similar image solely based on the same average color.
If I pick several icons on my desktop, for example:
And I will search for the last one (I'm still wondering what this file is):
You can clearly see what is most likely to be the match:
In different situations, this may not work. However when image size is given, it should be pretty reliable and lightning fast.
I can get a screenshot as BufferedImage object:
MSWindow window = MSWindow.windowFromName("Firefox", false);
BufferedImage img = window.screenshot();
//Or, if I can estimate smaller region for searching:
BufferedImage img2 = window.screenshotCrop(20,20,50,50);
Of course, the image to search image will be loaded from template saved in a file:
BufferedImage img = ImageIO.read(...whatever goes in there, I'm still confused...);
I explained what all I know so that we can focus on the only problem:
Q: How can I get average color on buffered image? How can I get such average color on sub-rectangle of that image?
Speed wins here. In this exceptional case, I consider it more valuable than code readability.
I think that no matter what you do, you are going to have an O(wh) operation, where w is your width and h is your height.
Therefore, I'm going to post this (naive) solution to fulfil the first part of your question as I do not believe there is a faster solution.
/*
* Where bi is your image, (x0,y0) is your upper left coordinate, and (w,h)
* are your width and height respectively
*/
public static Color averageColor(BufferedImage bi, int x0, int y0, int w,
int h) {
int x1 = x0 + w;
int y1 = y0 + h;
long sumr = 0, sumg = 0, sumb = 0;
for (int x = x0; x < x1; x++) {
for (int y = y0; y < y1; y++) {
Color pixel = new Color(bi.getRGB(x, y));
sumr += pixel.getRed();
sumg += pixel.getGreen();
sumb += pixel.getBlue();
}
}
int num = w * h;
return new Color(sumr / num, sumg / num, sumb / num);
}
There is a constant time method for finding the mean colour of a rectangular section of an image but it requires a linear preprocess. This should be fine in your case. This method can also be used to find the mean value of a rectangular prism in a 3d array or any higher dimensional analog of the problem. I will be using a gray scale example but this can be easily extended to 3 or more channels simply by repeating the process.
Lets say we have a 2 dimensional array of numbers we will call "img".
The first step is to generate a new array of the same dimensions where each element contains the sum of all values in the original image that lie within the rectangle that bounds that element and the top left element of the image.
You can use the following method to construct such an image in linear time:
int width = 1920;
int height = 1080;
//source data
int[] img = GrayScaleScreenCapture();
int[] helperImg = int[width * height]
for(int y = 0; y < height; ++y)
{
for(int x = 0; x < width; ++x)
{
int total = img[y * width + x];
if(x > 0)
{
//Add value from the pixel to the left in helperImg
total += helperImg[y * width + (x - 1)];
}
if(y > 0)
{
//Add value from the pixel above in helperImg
total += helperImg[(y - 1) * width + x];
}
if(x > 0 && y > 0)
{
//Subtract value from the pixel above and to the left in helperImg
total -= helperImg[(y - 1) * width + (x - 1)];
}
helperImg[y * width + x] = total;
}
}
Now we can use helperImg to find the total of all values within a given rectangle of img in constant time:
//Some Rectangle with corners (x0, y0), (x1, y0) , (x0, y1), (x1, y1)
int x0 = 50;
int x1 = 150;
int y0 = 25;
int y1 = 200;
int totalOfRect = helperImg[y1 * width + x1];
if(x0 > 0)
{
totalOfRect -= helperImg[y1 * width + (x0 - 1)];
}
if(y0 > 0)
{
totalOfRect -= helperImg[(y0 - 1) * width + x1];
}
if(x0 > 0 && y0 > 0)
{
totalOfRect += helperImg[(y0 - 1) * width + (x0 - 1)];
}
Finally, we simply divide totalOfRect by the area of the rectangle to get the mean value:
int rWidth = x1 - x0 + 1;
int rheight = y1 - y0 + 1;
int meanOfRect = totalOfRect / (rWidth * rHeight);
Here's a version based on k_g's answer for a full BufferedImage with adjustable sample precision (step).
public static Color getAverageColor(BufferedImage bi) {
int step = 5;
int sampled = 0;
long sumr = 0, sumg = 0, sumb = 0;
for (int x = 0; x < bi.getWidth(); x++) {
for (int y = 0; y < bi.getHeight(); y++) {
if (x % step == 0 && y % step == 0) {
Color pixel = new Color(bi.getRGB(x, y));
sumr += pixel.getRed();
sumg += pixel.getGreen();
sumb += pixel.getBlue();
sampled++;
}
}
}
int dim = bi.getWidth()*bi.getHeight();
// Log.info("step=" + step + " sampled " + sampled + " out of " + dim + " pixels (" + String.format("%.1f", (float)(100*sampled/dim)) + " %)");
return new Color(Math.round(sumr / sampled), Math.round(sumg / sampled), Math.round(sumb / sampled));
}