Removed benchmarking info to System.out.println
This commit is contained in:
parent
8c7513a056
commit
fa49592735
1 changed files with 5 additions and 7 deletions
|
@ -76,24 +76,22 @@ public class ImageFilterOp implements BufferedImageOp {
|
|||
// filters convert to integer internally anyhow
|
||||
cm = new SimpleColorModel();
|
||||
|
||||
// create a BufferedImage of only 1 pixel height for fetching the rows of the image in the correct format
|
||||
// This speeds up things by more than factor 2, compared with the standard BufferedImage.getRGB solution,
|
||||
// which is supposed to be fast too. This is probably the case because drawing to BufferedImages is
|
||||
// very optimized or even hardware accelerated.
|
||||
// create a BufferedImage of only 1 pixel height for fetching the rows of the image in the correct format (ARGB)
|
||||
// This speeds up things by more than factor 2, compared to the standard BufferedImage.getRGB solution,
|
||||
// which is supposed to be fast too. This is probably the case because drawing to BufferedImages uses
|
||||
// very optimized code which may even be hardware accelerated.
|
||||
BufferedImage row = new BufferedImage(width, 1, BufferedImage.TYPE_INT_ARGB);
|
||||
Graphics2D g2d = row.createGraphics();
|
||||
int pixels[] = ((DataBufferInt)row.getRaster().getDataBuffer()).getData();
|
||||
|
||||
// calculate scanline by scanline in order to safe memory.
|
||||
// It also seems to run faster like that
|
||||
long t = System.currentTimeMillis();
|
||||
for (int y = 0; y < height; y++) {
|
||||
g2d.drawImage(src, null, 0, -y);
|
||||
// now pixels contain the rgb values of the row y!
|
||||
// now pixels contains the rgb values of the row y!
|
||||
fltr.setPixels(0, y, width, 1, cm, pixels, 0, width);
|
||||
}
|
||||
g2d.dispose();
|
||||
System.out.println(System.currentTimeMillis() - t);
|
||||
|
||||
return consumer.getImage();
|
||||
}
|
||||
|
|
Loading…
Add table
Reference in a new issue