Develop Reference

Java rotation of pixel array

I have tried to make an algorithm in java to rotate a 2-d pixel array(Not restricted to 90 degrees), the only problem i have with this is: the end result leaves me with dots/holes within the image.
Here is the code :
for (int x = 0; x < width; x++)
{
for (int y = 0; y < height; y++)
{
int xp = (int) (nx + Math.cos(rotation) * (x - width / 2) + Math
.cos(rotation + Math.PI / 2) * (y - height / 2));
int yp = (int) (ny + Math.sin(rotation) * (x - width / 2) + Math
.sin(rotation + Math.PI / 2) * (y - height / 2));
int pixel = pixels[x + y * width];
Main.pixels[xp + yp * Main.WIDTH] = pixel;
}
}
'Main.pixels' is an array connected to a canvas display, this is what is displayed onto the monitor.
'pixels' and the function itself, is within a sprite class. The sprite class grabs the pixels from a '.png' image at initialization of the program.
I've tried looking at the 'Rotation Matrix' solutions. But they are too complicated for me. I have noticed that when the image gets closer to a point of 45 degrees, the image is some-what stretched ? What is going wrong? And what is the correct code; that adds the pixels to a larger scale array(E.g. Main.pixels[]).
Needs to be java! and relative to the code format above. I am not looking for complex examples, simply because i will not understand(As said above). Simple and straight to the point, is what i am looking for.
How id like the question to be answered.
Your formula is wrong because ....
Do this and the effect will be...
Simplify this...
Id recommend...
Im sorry if im asking to much, but i have looked for an answer relative to this question, that i can understand and use. But to always either be given a rotation of 90 degrees, or an example from another programming language.

You are pushing the pixels forward, and not every pixel is hit by the discretized rotation map. You can get rid of the gaps by calculating the source of each pixel instead.
Instead of
for each pixel p in the source
pixel q = rotate(p, theta)
q.setColor(p.getColor())
try
for each pixel q in the image
pixel p = rotate(q, -theta)
q.setColor(p.getColor())
This will still have visual artifacts. You can improve on this by interpolating instead of rounding the coordinates of the source pixel p to integer values.
Edit: Your rotation formulas looked odd, but they appear ok after using trig identities like cos(r+pi/2) = -sin(r) and sin(r+pi/2)=cos(r). They should not be the cause of any stretching.

To avoid holes you can:
compute the source coordinate from destination
(just reverse the computation to your current state) it is the same as Douglas Zare answer
use bilinear or better filtering
use less then single pixel step
usually 0.75 pixel is enough for covering the holes but you need to use floats instead of ints which sometimes is not possible (due to performance and or missing implementation or other reasons)
Distortion
if your image get distorted then you do not have aspect ratio correctly applied so x-pixel size is different then y-pixel size. You need to add scale to one axis so it matches the device/transforms applied. Here few hints:
Is the source image and destination image separate (not in place)? so Main.pixels and pixels are not the same thing... otherwise you are overwriting some pixels before their usage which could be another cause of distortion.
Just have realized you have cos,cos and sin,sin in rotation formula which is non standard and may be you got the angle delta wrongly signed somewhere so
Just to be sure here an example of the bullet #1. (reverse) with standard rotation formula (C++):
float c=Math.cos(-rotation);
float s=Math.sin(-rotation);
int x0=Main.width/2;
int y0=Main.height/2;
int x1= width/2;
int y1= height/2;
for (int a=0,y=0; y < Main.height; y++)
for (int x=0; x < Main.width; x++,a++)
{
// coordinate inside dst image rotation center biased
int xp=x-x0;
int yp=y-y0;
// rotate inverse
int xx=int(float(float(xp)*c-float(yp)*s));
int yy=int(float(float(xp)*s+float(yp)*c));
// coordinate inside src image
xp=xx+x1;
yp=yy+y1;
if ((xp>=0)&&(xp<width)&&(yp>=0)&&(yp<height))
Main.pixels[a]=pixels[xp + yp*width]; // copy pixel
else Main.pixels[a]=0; // out of src range pixel is black
}

Two Shape objects appear to be intersecting at the wrong location?

I am trying to create a Shape with the centre of the ship being in the middle.
one.x and one.z is the X and Z positions of the ship. The ship size is about 100 on the X, and 50 on the Z.
Shape my = new Rectangle(
(int) one.x - disToLeft, // upper-left corner X
(int) one.z - disToTop, // upper-left corner Y
disToLeft + disToRight, // width
disToTop + disToBottom // height
);
I'm then rotating the Shape, to of course be facing the correct way. This appears to work:
int rectWidth = (disToLeft + disToRight);
int rectHeight = (disToTop + disToBottom);
AffineTransform tr = new AffineTransform();
// rotating in central axis
tr.rotate(
Math.toRadians(one.rotation),
x + (disToLeft + disToRight) / 2,
z + (disToTop + disToBottom) / 2
);
my = tr.createTransformedShape(my);
I am then doing the exact same thing with another Shape, and testing for intersection. This works.
However, it feels like the Shape is the incorrect dimensions. Or something. My ship is colliding very far out to one side (outside where it graphically exists), but through the other side, I can almost go right through the ship before any collision is detected!
Basically the Shapes are simply intersecting at the wrong location. And I cannot work out why. Either the shape, the location, or the rotation must be wrong.
int disToLeft = 100;
int disToRight = 100;
int disToTop = 150;
int disToBottom = 100;
These are the distance from the centre to the left, right, top, and bottom sides.
I am using Z instead of Y because my game is in a 3D world and the sea level is pretty much the same (hence I don't need to worry about Y axis collision!).
Update:
After doing a lot of testing, I have discovered that the top of the rectangle is in the middle! I have done a lot of messing around, but without being able to graphically see the squares, it's been very hard to test.
This means that the box is on the side of the ship, like this:
Obviously when the ship on the left rotates to what it's like in this picture, a collision is detected.

It seems that your rotation is wrong. From my understanding of math it should be
tr.rotate(Math.toRadians(one.rotation), x + (disToRight - disToLeft) /2, z + (disToBottom - disToTop) /2);
Note the signs and the order of the variables
Edit:
Let's take apart the formula:
Your Rectangle is defined like this:
x-coordinate (x): one.x-disToLeft
y-coordinate (y): one.z-disToTop
width: disToLeft+disToRight
height: disToTop+disToBottom
The centre of the Rectangle (where you are rotating) is therefore:
(x+width/2)
(y+height/2)
if you replace x, width, y and height with the declarations above you get
(one.x-disToLeft + (disToLeft+disToRight)/2)
(one.z-disToTop + (disToTop+disToBottom)/2)
This is already the point you need, but it can be simplyfied:
one.x- disToLeft + (disToLeft+disToRight)/2
is equal to
one.x-(2*disToLeft/2)+(disToLeft/2)+(disToRight/2)
is equal to
one.x-(distoLeft/2) + (disToRight/2)
is equal to
one.x+(disToRight-disToLeft)/2
The other coordinate works exactly the same.

Java: Rotate a Point around an other using Google Maps Coordinates

From Google Earth I got a Box with coordinates for a picture, like following:
<LatLonBox>
<north>53.10685</north>
<south>53.10637222222223</south>
<east>8.853144444444444</east>
<west>8.851858333333333</west>
<rotation>-26.3448</rotation>
</LatLonBox>
Now I want to test weather a point intersect with this LatLonBox.
My base idea to check, whether a point intersect with the LatLonBox was, to rotate the point back by the given angle, and then to test whether the point intersect with a regular (not rotated) rectangle.
I tried to calculate the rotation manually:
public static MyGeoPoint rotatePoint(MyGeoPoint point, MyGeoPoint origion, double degree)
{
double x = origion.getLatitude() + (Math.cos(Math.toRadians(degree)) * (point.getLatitude() - origion.getLatitude()) - Math.sin(Math.toRadians(degree)) * (point.getLongitude() - origion.getLongitude()));
double y = origion.getLongitude() + (Math.sin(Math.toRadians(degree)) * (point.getLatitude() - origion.getLatitude()) + Math.cos(Math.toRadians(degree)) * (point.getLongitude() - origion.getLongitude()));
return new MyGeoPoint(x, y);
}
public boolean intersect(MyGeoPoint geoPoint)
{
geoPoint = MyGeoPoint.rotatePoint(geoPoint, this.getCenter(), - this.getRotation());
return (geoPoint.getLatitude() < getTopLeftLatitude()
&& geoPoint.getLatitude() > getBottomRightLatitude()
&& geoPoint.getLongitude() > getTopLeftLongitude()
&& geoPoint.getLongitude() < getBottomRightLongitude());
}
And it seems that the results are wrong.
LatLonBox box = new LatLonBox(53.10685, 8.851858333333333, 53.10637222222223, 8.853144444444444, -26.3448);
MyGeoPoint point1 = new MyGeoPoint(53.106872, 8.852311);
MyGeoPoint point2 = new MyGeoPoint(53.10670378322918, 8.852967186822669);
MyGeoPoint point3 = new MyGeoPoint(53.10652664993972, 8.851994565566875);
MyGeoPoint point4 = new MyGeoPoint(53.10631650700605, 8.85270995172055);
System.out.println(box.intersect(point1));
System.out.println(box.intersect(point2));
System.out.println(box.intersect(point3));
System.out.println(box.intersect(point4));
The result is true, false, false, true. But it should be 4x true.
Probably I´, making some kind of error in reasoning.
Maybe because the latitude values are getting bigger upwards. But I don´t knwo how to change the formular.
I need some help ...
EDIT:
I think my basic idea and formular is right. Also I found similar solutions eg. link and couldn´t find any difference.
So I think the only possible error source is, that the axis are not proportional. So the problem is how to take account of this.
I hope someone has got an idea.

The problem was indeed that the axis were not proportional.
The following method takes care of it.
public static MyGeoPoint rotatePoint(MyGeoPoint point, MyGeoPoint origion, double degree)
{
double x = origion.longitude + (Math.cos(Math.toRadians(degree)) * (point.longitude - origion.longitude) - Math.sin(Math.toRadians(degree)) * (point.latitude - origion.latitude) / Math.abs(Math.cos(Math.toRadians(origion.latitude)));
double y = origion.latitude + (Math.sin(Math.toRadians(degree)) * (point.longitude - origion.longitude) * Math.abs(Math.cos(Math.toRadians(origion.latitude))) + Math.cos(Math.toRadians(degree)) * (point.latitude - origion.latitude));
return new MyGeoPoint(x, y);
}

if I understand correctly you want to check if these four points are in rotated rectangle.
I would recommend checking not by corner points because your rectangle is rotated but:
if you have rotated rectangle ABCD then calculate lines |AB|, |BC|,|CD| and |DA|. If you have two points then use y=ax+b (you will calculate a,b by by giving [x,y] of both coordinates that gives you two easy equatations).
Finally function intersect will check
if point <= line |CD|
AND point >= line |AB|
AND point <= line |BC|
AND point >= |DA|
then it is inside rect.
This can be done when your point P[x,y] you put in ax+y+b (a>0 or -ax-y-b). If it is zero it is lying on the line, if it is < than it is under line or "on the left side". Hope I helped..
BTW why are you using -degree value, which you multiply by -1 , is it necessary?

The problem appears to be that the data structure LatLonBox doesn't make any sense as a description for the boundary of a picture. A box in lat-lon coordinates is not a geometric rectangle. (Think about a box near or including the north pole.) You need to re-think your application to deal in a lat/lon coordinate for the center of the picture and then deal with the rotation as an angle with respect to lines of latitude (parallel to the equator). (Even then, a picture with center on the north or south pole will be a degenerate case that must be handled separately.) So a box should properly be something like:
<geobox>
<center_lat>41</center_lat>
<center_lon>-74</center_lon>
<rotation_degrees_ccw>-23</rotation_degrees_ccw>
<width>1000</width> <!-- in pixels or meters, but not in degrees! -->
<height>600</height> <!-- same as above -->
</geobox>
Having said all that, suppose you have a true geometric box centered at (x0,y0), width w, height h, rotated angle T about its center. Then you can test a point P(x,y) for membership in the box with the following. You need the transformation that takes the box to the origin and aligns it with the axes. This is Translate(-x0,-y0) then Rotate(-T). This transformation as a matrix is
[cos(-T) -sin(-T) 0][1 0 -x0] [ cos(T) sin(T) -x0*cos(T)-y0*sin(T)]
[sin(-T) cos(-T) 0][0 1 -y0] = [-sin(T) cos(T) x0*sin(T)-y0*cos(T)]
[0 0 1][0 0 1] [ 0 0 1 ]
You want to apply this transformation to the point to be tested and then see if it lies in the desired box:
// Transform the point to be tested.
ct = cos(T);
st = sin(T);
xp = ct * x + st * y - x0 * ct - y0 * st;
yp = -st * x + ct * y + x0 * st - y0 * ct;
// Test for membership in the box.
boolean inside = xp >= -w/2 && xp <= w/2 && yp >= -h/2 && yp <= h/2;
It's late and I haven't checked this arithmetic, but it's close. Say if it doesn't work.

Line detection | Angle detection with Java

I'm processing some images that my UGV (Unmanned Ground Vehichle) captures to make it move on a line.
I want to get the angle of that line based on the horizon. I'll try to explain with a few examples:
The image above would make my UGV to keep straight ahead, as the angle is about 90 degrees.
But the following would make it turn left, as the angle compaired to the horizon rounds about 120.
I could successfully transform those images into the image below using otsu for thresholding:
And also used an edge detection algorithm to get this:
But I'm stuck right now trying to find an algorithm that detecs those edges/lines and outputs - or helps me to output - the angle of such line..

Here's my attempt using ImageJ:
// Open the Image
ImagePlus image = new ImagePlus(filename);
// Make the Image 8 bit
IJ.run(image, "8-bit", "");
// Apply a Threshold (0 - 50)
ByteProcessor tempBP = (ByteProcessor)image.getProcessor();
tempBP.setThreshold(0, 50, 0);
IJ.run(image, "Convert to Mask", "");
// Analyze the Particles
ParticleAnalyzer pa = new ParticleAnalyzer(
ParticleAnalyzer.SHOW_MASKS +
ParticleAnalyzer.IN_SITU_SHOW,
1023 +
ParticleAnalyzer.ELLIPSE
, rt, 0.0, 999999999, 0, 0.5);
IJ.run(image, "Set Measurements...", "bounding fit redirect=None decimal=3");
pa.analyze(image);
int k = 0;
double maxSize = -1;
for (int i = 0; i < rt.getCounter(); i ++) {
// Determine creteria for best oval.
// The major axis should be much longer than the minor axis.
// let k = best oval
}
double bx = rt.getValue("BX", k);
double by = rt.getValue("BY", k);
double width = rt.getValue("Width", k);
double height = rt.getValue("Height", k);
// Your angle:
double angle = rt.getValue("Angle", k);
double majorAxis = rt.getValue("Major", k);
double minorAxis = rt.getValue("Minor", k);
How the code works:
Make the image grayscaled.
Apply a threshold on it to only get the dark areas. This assumes the lines will always be near black.
Apply a Particle Analyzer to find Ellipses on the image.
Loop through the "Particles" to find ones that fit our criteria.
Get the angle from our Particle.
Here's an example of what the image looks like when I analyze it:
NOTE: The code is untested. I just converted what I did in the Visual ImageJ into Java.

AndEngine - Getting Center Coordinates of a Sprite

I have a sprite in my game and I want to get it's center coordinates. The sprite can be rotated though so I can't just get it's x/y coordinates then add half of the sprite's width/height because the getWidth and getHeight methods return the dimensions of the original, unrotated sprite.
I tried getSceneCenterCoordinates() but that for some reason returns the same coordinates for all sprites even if they are nowhere near each other.
Here's a graphic to describe what I want, the red dot is the coordinate I want, and the width/height labels on the right side figure respresent what I WANT the getWidth/Height methods to return (but they don't):

You can use coordinates transformer
final float[] spriteCoordinates = sprite.convertLocalToSceneCoordinates(x,y);
final float canonX = spriteCoordinates[VERTEX_INDEX_X];
final float canonY = spriteCoordinates[VERTEX_INDEX_Y];
Sprite sprite_in_fixedpoint_of_the_first_sprite=new Sprite(canonX,canonY,textureregion)

Get the minimum and maximum value of all the corners (for both - x and y components).
Then take average from max and min to get middle:
middleX = (maxX + minX) / 2
Repeat the same for y component.

Consider something like this
Rectangle test = new Rectangle(100, 100, 50, 100, vboManager);
mainScene.attachChild(test);
System.out.println("Before RotCtrX = " + test.getRotationCenterX());
System.out.println("Before RotCtrY = " + test.getRotationCenterY());
test.setRotation(45);
System.out.println("After RotCtrX = " + test.getRotationCenterX());
System.out.println("After RotCtrY = " + test.getRotationCenterY());
and the result are
System.out(4526): Before RotCtrX = 25.0
System.out(4526): Before RotCtrY = 50.0
System.out(4526): After RotCtrX = 25.0
System.out(4526): After RotCtrY = 50.0
AndEngine Entities rotate around the center (unless you change the RotationCenter values), so applying setRotate() will not affect the "center" point location
those "center" points are relative to the Entity, so if you need the actual screen coordinates, you will need to add those to the getX() and getY() values - which BTW also won't change based solely on applying a setRotate()

I figured out how to use getSceneCenterCoordinates() properly, which is to hand it a float[]. Here is my solution:
float[] objCenterPos = new float[2];
obj.sprite.getSceneCenterCoordinates(objCenterPos);
Log.d(this.toString(), "Coordinates: ("+objCenterPos[0]+","+objCenterPos[1]+")");