Am 03.10.2018 um 21:14 schrieb Daniel Persson:
Hi Tilman
Thank you for looking into this for me, I've run your code with different
java versions and I can see the exact opposite time metrics.
In my initial test Java 8 was fastest, Java 9 was slowest and 10-11 was
somewhere in the middle.
Maybe different color profiles are good at different images?
Do you have any more suggestions?
Just a guess, but maybe you are comparing apples and oranges? It looks like
Daniel is using Linux and AFAIK Tilman is using Windows. That could be the
reason for the different results.
Andreas
I'll report these findings to Oracle tomorrow.
Best regards
Daniel
$ java -jar target/ColorConvTest-1.0-SNAPSHOT-jar-with-dependencies.jar
Raster: ByteInterleavedRaster: width = 2577 height = 1540 #numDataElements
4 dataOff[0] = 0
60178
$ /opt/openjdk/jdk-9.0.4/bin/java -jar
target/ColorConvTest-1.0-SNAPSHOT-jar-with-dependencies.jar
Raster: ByteInterleavedRaster: width = 2577 height = 1540 #numDataElements
4 dataOff[0] = 0
33327
$ /opt/openjdk/jdk-10.0.2/bin/java -jar
target/ColorConvTest-1.0-SNAPSHOT-jar-with-dependencies.jar
Raster: ByteInterleavedRaster: width = 2577 height = 1540 #numDataElements
4 dataOff[0] = 0
56596
$ /opt/openjdk/jdk-11/bin/java -jar
target/ColorConvTest-1.0-SNAPSHOT-jar-with-dependencies.jar
Raster: ByteInterleavedRaster: width = 2577 height = 1540 #numDataElements
4 dataOff[0] = 0
56988
On Wed, Oct 3, 2018 at 8:31 PM Tilman Hausherr <[email protected]>
wrote:
Am 03.10.2018 um 20:08 schrieb Tilman Hausherr:
I could expand that one later so that it uses the raster of the actual
JPEG image from your PDF.
Your JPEG file is here:
https://drive.google.com/file/d/1HvXMet6Wygzmko9ZmY9_zlvhv72v_QHf/view?usp=sharing
The modified code - please test on different openjdks. It took 137935 on
oracle jdk8. (you may want to change the "50")
import java.awt.*;
import java.awt.color.*;
import java.awt.image.*;
import java.io.FileInputStream;
import javax.imageio.ImageIO;
import javax.imageio.ImageReadParam;
import javax.imageio.ImageReader;
import javax.imageio.stream.ImageInputStream;
import javax.imageio.stream.MemoryCacheImageInputStream;
public class CConv
{
static BufferedImage toRGBImageAWT(WritableRaster raster,
ColorSpace colorSpace)
{
ColorModel colorModel = new ComponentColorModel(colorSpace,
false, false, Transparency.OPAQUE,
raster.getDataBuffer().getDataType());
BufferedImage src = new BufferedImage(colorModel, raster,
false, null);
BufferedImage dest = new BufferedImage(raster.getWidth(),
raster.getHeight(),
BufferedImage.TYPE_INT_RGB);
ColorConvertOp op = new ColorConvertOp(null);
//System.out.println("src="+src);
//System.out.println("dst="+dest);
op.filter(src, dest);
return dest;
}
public static void main(String args[]) throws Exception
{
ImageReader reader =
ImageIO.getImageReadersByFormatName("JPEG").next();
WritableRaster raster;
FileInputStream fis = new FileInputStream("cmyk.jpg");
try (ImageInputStream iis = new MemoryCacheImageInputStream(fis))
{
reader.setInput(iis);
ImageReadParam irp = reader.getDefaultReadParam();
raster = (WritableRaster) reader.readRaster(0, irp);
}
System.out.println("Raster: " + raster);
ICC_Profile profile =
ICC_Profile.getInstance("ISOcoated_v2_300_bas.icc");
ICC_ColorSpace colorSpace = new ICC_ColorSpace(profile);
long t0 = System.currentTimeMillis();
for (int i = 0; i < 50; i++)
{
toRGBImageAWT(raster, colorSpace);
}
long t1 = System.currentTimeMillis();
System.out.println((t1 - t0));
}
}
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
