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
}

unable to calculate itext PdfPTable/PdfPCell height properly

I'm facing a problem while trying to generate a PdfPTable and calculate its height before adding it to a document. The method calculateHeights of PdfPTable returned the height a lot greater than the height of a page (while the table is about 1/4 of page's height), so I wrote a method to calculate the height:
protected Float getVerticalSize() throws DocumentException, ParseException, IOException {
float overallHeight=0.0f;
for(PdfPRow curRow : this.getPdfObject().getRows()) {
float maxHeight = 0.0f;
for(PdfPCell curCell : curRow.getCells()) {
if(curCell.getHeight()>maxHeight) maxHeight=curCell.getHeight();
}
overallHeight+=maxHeight;
}
return overallHeight;
}
where getPdfObject method returns a PdfPTable object.
Using debugger I've discovered that lly and ury coordinate difference (and thus the height) of cell's rectangle is much bigger than it looks after adding a table to a document (for example, one cell is 20 and the other is 38 height while they look like the same on a page). There is nothing in the cell except a paragraph with a chunk in it:
Font f = getFont();
if (f != null) {
int[] color = getTextColor();
if(color != null) f.setColor(color[0],color[1],color[2]);
ch = new Chunk(celltext, f);
par = new Paragraph(ch);
}
cell = new PdfPCell(par);
cell.setHorizontalAlignment(getHorizontalTextAlignment());
cell.setVerticalAlignment(getVerticalTextAlignment());
A table then has a cell added and setWidthPercentage attribute set to a some float.
What am I doing wrong? Why does cell's proportions are different from those I see after generating PDF? Maybe I'm calculating the height wrong? Isn't it the height of a cell on a PDF page should strictly be the difference between lly and ury coordinates
Sorry I haven't shown the exact code, because the PDF is being generated of XML using lots of intermediate steps and objects and it is not very useful "as is" I guess...
Thanks in advance!

The height of table added to a page where the available width is 400 is different from the height of a table added to a page where the available width is 1000. There is no way you can measure the height correctly until the width is defined.
Defining the width can be done by adding the table to the document. Once the table is rendered, the total height is known.
If you want to know the height in advance, you need to define the width in advance. For instance by using:
table.setTotalWidth(400);
table.setLockedWidth(true);
This is explained in the TableHeight example. In table_height.pdf, you see that iText returns a height of 0 before adding a table and a height of 48 after adding the table. iText initially returns 0 because there is no way to determine the actual height.
We then take the same table and we define a total width of 50 (which is much smaller than the original 80% of the available width on the page). Now when we calculate the height of the table with the same contents, iText returns 192 instead of 48. When you look at the table on the page, the cause of the difference in height is obvious.

Inorder to get dynamic table height we should set and lock width of table.
Here, 595 is A4 size paper width.
table.setTotalWidth(595);
table.setLockedWidth(true);

Apache POI: Word get picture dimensions

I am trying to determine the actual size of embedded images in MS Word documents using
HWPFDocument docB = new HWPFDocument(fileInputStream);
PicturesTable picB = docB.getPicturesTable();
List picturesB = picB.getAllPictures();
for (Object o : picturesB) {
Picture pic = (Picture) o;
int height = pic.getHeight();
int width = pic.getWidht();
}
This works fine for some images but for others getHeight() and getWidth() will only return -1 as stated in the documentation.
So is there any other way to get the actual size of these pictures in the document?

Instead of getHeight() and getWidth() you may resort to getDxaGoal(), resp. getDyaGoal(). These represent the original size of the image prior to scaling / cropping in twips.
You can then multiply this with getVerticalScalingFactor() / 1000.0 resp. getHorizontalScalingFactor() / 1000.0 to get the final (rendered) size. Division by 1000.0 is necessary because those scaling factors are given per mille.

Pixels that text will give

Would it be possible to get the width of a string based on its font?
So for example, if the font size is 40, and the string was "hi", could you do 2*40.
An example of this would be
startX = 260; startWidth = "Start".length()*getFont().getSize();
startY = getHeight()-startWidth-20;
startHeight = getFont().getSize();

It really depends on the type of font, whether it's monospaced or proportional. Most font now are proportional so you would not simply be able to calculate num_of_chars * n.
The gui interface you are using should give you the width of the container once it has been populated with the given text.

Java: Getting a font with a specific height in pixels

It’s easy to determine the rendered height of a font using FontMetrics, but what about the other way around? How can I obtain a font that will fit into a specific height in pixels?
"Give me Verdana in a size that is 30 pixels high from ascender to descender."
How do I ask Java for this?

I know this is a very old question, but someone might still find it:
The font height in Java (and many other places) is given in "typographic points", which are defined as roughly 1/72nd of an inch.
To calculate the points needed for a certain pixel height, you should be able to use the following:
double fontSize= 72.0 * pixelSize / Toolkit.getDefaultToolkit().getScreenResolution();
I haven't tested this extensively yet, but it seems to work for the monitors that I've used. I'll report back if I ever find a case where it doesn't work.
For the standard system fonts I've used this with, this sets the height of a capital letter (i.e. the ascent) to the provided pixel size. If you need to set the ascent+descent to the pixel size, you can correct the value using the FontMetrics:
FontMetrics m= g.getFontMetrics(font); // g is your current Graphics object
double totalSize= fontSize * (m.getAscent() + m.getDescent()) / m.getAscent();
Of course, the actual pixel-height of some specific letters will depend on the letter and the font used, so if you want to make sure that your "H" is some exact number of pixels tall, you might still want to use the trial-and-error methods mentioned in the other answers. Just keep in mind that if you use these methods to get the size for each specific text you want to display (as #Bob suggested), you might end up with a random font-size-mess on your screen where a text like "ace" will have much bigger letters than "Tag". To avoid this, I would pick one specific letter or letter sequence ("T" or "Tg" or something) and fix that one to your pixel height once and then use the font size you get from that everywhere.

I don't think there's a "direct" way to find a font by height; only an indirect way... by looping through the sizes, and testing the height of each is <= required height.
If you're doing this once, just loop through them... if you've doing it "on the fly" then do a binary search, it'll be quicker.

I'm not aware of a way to get a font by its actual height in pixels. It depends on the context it's used in so there's probably no shorter way than to sample for the best match. It should be pretty quick to look for sizes up or down from the designed height. Here's an example method that does that:
public Font getFont(String name, int style, int height) {
int size = height;
Boolean up = null;
while (true) {
Font font = new Font(name, style, size);
int testHeight = getFontMetrics(font).getHeight();
if (testHeight < height && up != Boolean.FALSE) {
size++;
up = Boolean.TRUE;
} else if (testHeight > height && up != Boolean.TRUE) {
size--;
up = Boolean.FALSE;
} else {
return font;
}
}
}

WhiteFang34's code is useful in combination with the following method that returns the actual height of a specific string. It might be a bit slow for real-time rendering, especially for large fonts/strings and I'm sure it can be further optimised, but for now it meets my own needs and is fast enough to run in a back-end process.
/*
* getFontRenderedHeight
* *************************************************************************
* Summary: Font metrics do not give an accurate measurement of the rendered
* font height for certain strings because the space between the ascender
* limit and baseline is not always fully used and descenders may not be
* present. for example the strings '0' 'a' 'f' and 'j' are all different
* heights from top to bottom but the metrics returned are always the same.
* If you want to place text that exactly fills a specific height, you need
* to work out what the exact height is for the specific string. This method
* achieves that by rendering the text and then scanning the top and bottom
* rows until the real height of the string is found.
*/
/**
* Calculate the actual height of rendered text for a specific string more
* accurately than metrics when ascenders and descenders may not be present
* <p>
* Note: this method is probably not very efficient for repeated measurement
* of large strings and large font sizes but it works quite effectively for
* short strings. Consider measuring a subset of your string value. Also
* beware of measuring symbols such as '-' and '.' the results may be
* unexpected!
*
* #param string
* The text to measure. You might be able to speed this process
* up by only measuring a single character or subset of your
* string i.e if you know your string ONLY contains numbers and
* all the numbers in the font are the same height, just pass in
* a single digit rather than the whole numeric string.
* #param font
* The font being used. Obviously the size of the font affects
* the result
* #param targetGraphicsContext
* The graphics context the text will actually be rendered in.
* This is passed in so the rendering options for anti-aliasing
* can be matched.
* #return Integer - the exact actual height of the text.
* #author Robert Heritage [mrheritage#gmail.com]
*/
public Integer getFontRenderedHeight(String string, Font font, Graphics2D targetGraphicsContext) {
BufferedImage image;
Graphics2D g;
Color textColour = Color.white;
// In the first instance; use a temporary BufferedImage object to render
// the text and get the font metrics.
image = new BufferedImage(1, 1, BufferedImage.TYPE_INT_RGB);
g = image.createGraphics();
FontMetrics metrics = g.getFontMetrics(font);
Rectangle2D rect = metrics.getStringBounds(string, g);
// now set up the buffered Image with a canvas size slightly larger than
// the font metrics - this guarantees that there is at least one row of
// black pixels at the top and the bottom
image = new BufferedImage((int) rect.getWidth() + 1, (int) metrics.getHeight() + 2, BufferedImage.TYPE_INT_RGB);
g = image.createGraphics();
// take the rendering hints from the target graphics context to ensure
// the results are accurate.
g.setRenderingHint(RenderingHints.KEY_ANTIALIASING, targetGraphicsContext.getRenderingHint(RenderingHints.KEY_ANTIALIASING));
g.setRenderingHint(RenderingHints.KEY_TEXT_ANTIALIASING, targetGraphicsContext.getRenderingHint(RenderingHints.KEY_TEXT_ANTIALIASING));
g.setColor(textColour);
g.setFont(font);
g.drawString(string, 0, image.getHeight());
// scan the bottom row - descenders will be cropped initially, so the
// text will need to be moved up (down in the co-ordinates system) to
// fit it in the canvas if it contains any. This may need to be done a
// few times until there is a row of black pixels at the bottom.
boolean foundBottom, foundTop = false;
int offset = 0;
do {
g.setColor(Color.BLACK);
g.fillRect(0, 0, image.getWidth(), image.getHeight());
g.setColor(textColour);
g.drawString(string, 0, image.getHeight() - offset);
foundBottom = true;
for (int x = 0; x < image.getWidth(); x++) {
if (image.getRGB(x, image.getHeight() - 1) != Color.BLACK.getRGB()) {
foundBottom = false;
}
}
offset++;
} while (!foundBottom);
System.out.println(image.getHeight());
// Scan the top of the image downwards one line at a time until it
// contains a non-black pixel. This loop uses the break statement to
// stop the while loop as soon as a non-black pixel is found, this
// avoids the need to scan the rest of the line
int y = 0;
do {
for (int x = 0; x < image.getWidth(); x++) {
if (image.getRGB(x, y) != Color.BLACK.getRGB()) {
foundTop = true;
break;
}
}
y++;
} while (!foundTop);
return image.getHeight() - y;
}