Develop Reference

Grayscale PNG writing from BufferedImage using PNGJ - hightest-level approach

I just want to, at first, use PNGJ at the highest possible level to write a grayscale PNG with bit-depth 8.
I am working from a BufferedImage. Here's a snippet of the code used:
BufferedImage scaledBI;
ByteArrayOutputStream baos = new ByteArrayOutputStream();
...
ImageInfo imi = new ImageInfo(scaledBI.getWidth(), scaledBI.getHeight(), 8, false, true, false);
PngWriter pngw = new PngWriter(baos, imi);
pngw.setCompLevel(3);
DataBufferByte db =((DataBufferByte) scaledBI.getRaster().getDataBuffer());
byte[] dbbuf = db.getData();
ImageLineByte line = new ImageLineByte(imi, dbbuf);
for (int row = 0; row < imi.rows; row++) {
pngw.writeRow(line, row);
}
pngw.end();
return baos.toByteArray();
The original image is in j2k and is a palm print scan. The PNG looks similar to the background of the j2k with some vertical gray lines of different shades.
I tried the same buffered image data with ImageIO:
ImageIO.write(scaledBI, "png", baos);
and com.pngencoder.PngEncoder:
new PngEncoder().withBufferedImage(scaledBI).withCompressionLevel(3).toStream(baos);
which both render the image properly.
The long term aim is to improve on the speed offered by ImageIO and PngEncoder (which doesn't support grayscale).
I am stuck on just trying to run PNGJ with all default settings. I looked at the snippets in PNGJ Wiki Snippets and the test code but I couldn't find a simple example with grayscale without scanline or chunk manipulation.
Looking at the PNGJ code it seems like we are properly going through block deflation row after row. What am I missing here?

my problem sprung from my misunderstanding of what ImageLineByte stores, although as the name suggests it deals with a line, not the entire image data. I got confused by the fact that it references the image info. So my output was fine once I used ImageByteLine for each line:
for (int row = 0, offset = 0; row < imi.rows; row++) {
int newOffset = offset + imi.cols;
byte[] lineBuffer = Arrays.copyOfRange(dbbuf, offset, newOffset);
ImageLineByte line = new ImageLineByte(imi, lineBuffer);
pngw.writeRow(line, row);
offset = newOffset;
}
Sorry for the trouble.

Writing javafx.scene.image.Image to file?

How would I go about writing a javafx.scene.image.Image image to a file. I know you can use ImageIO on BufferedImages but is there any way to do it with a javafx Image?

Just convert it to a BufferedImage first, using javafx.embed.swing.SwingFXUtils:
Image image = ... ; // javafx.scene.image.Image
String format = ... ;
File file = ... ;
ImageIO.write(SwingFXUtils.fromFXImage(image, null), format, file);

Almost 3 years later and I now have the knowledge to do and answer this. Yes the original answer was also valid but it involved first converting the image to a BufferedImage and I ideally wanted to avoid swing entirely. While this does output the raw RGBA version of the image that's good enough for what I needed to do. I actually could just use raw BGRA since I was writing the software to open the result but since gimp can't open that I figure I'd convert it to RGBA.
Image img = new Image("file:test.png");
int width = (int) img.getWidth();
int height = (int) img.getHeight();
PixelReader reader = img.getPixelReader();
byte[] buffer = new byte[width * height * 4];
WritablePixelFormat<ByteBuffer> format = PixelFormat.getByteBgraInstance();
reader.getPixels(0, 0, width, height, format, buffer, 0, width * 4);
try {
BufferedOutputStream out = new BufferedOutputStream(new FileOutputStream("test.data"));
for(int count = 0; count < buffer.length; count += 4) {
out.write(buffer[count + 2]);
out.write(buffer[count + 1]);
out.write(buffer[count]);
out.write(buffer[count + 3]);
}
out.flush();
out.close();
} catch(IOException e) {
e.printStackTrace();
}

JavaFX has no built-in method to do this.
To solve this problem, I implemented a very small (< 20KiB) library for writing PNG files: https://github.com/Glavo/SimplePNG
Usage:
Image img = new Image("path-to-image.jpg");
try (PNGWriter writer = new PNGWriter(Files.newOutputStream(Path.of("output.png")))) {
writer.write(PNGJavaFXUtils.asArgbImage(img));
}
// Or you can use the shortcut:
// PNGJavaFXUtils.writeImage(img, Path.of("output.png"));
It has no dependencies and can work on the JRE that only have java.base.
I avoid the dependence on Java AWT (java.desktop) through it.

Java: saving image as JPEG skew problem

I am trying to save an image to JPEG. The code below works fine when image width is a multiple of 4, but the image is skewed otherwise. It has something to do with padding. When I was debugging I was able to save the image as a bitmap correctly, by padding each row with 0s. However, this did not work out with the JPEG.
Main point to remember is my image is represented as bgr (blue green red 1 byte each) byte array which I receive from a native call.
byte[] data = captureImage(OpenGLCanvas.getLastFocused().getViewId(), x, y);
if (data.length != 3*x*y)
{
// 3 bytes per pixel
return false;
}
// create buffered image from raw data
DataBufferByte buffer = new DataBufferByte(data, 3*x*y);
ComponentSampleModel csm = new ComponentSampleModel(DataBuffer.TYPE_BYTE, x, y, 3, 3*x, new int[]{0,1,2} );
WritableRaster raster = Raster.createWritableRaster(csm, buffer, new Point(0,0));
BufferedImage buff_image = new BufferedImage(x, y, BufferedImage.TYPE_INT_BGR); // because windows goes the wrong way...
buff_image.setData(raster);
//save the BufferedImage as a jpeg
try
{
File file = new File(file_name);
FileOutputStream out = new FileOutputStream(file);
JPEGImageEncoder encoder = JPEGCodec.createJPEGEncoder(out);
JPEGEncodeParam param = encoder.getDefaultJPEGEncodeParam(buff_image);
param.setQuality(1.0f, false);
encoder.setJPEGEncodeParam(param);
encoder.encode(buff_image);
out.close();
// or JDK 1.4
// ImageIO.write(image, "JPEG", out);
}
catch (Exception ex)
{
// Write permissions on "file_name"
return false;
}
I also looked on creating the JPEG in C++ but there was even less material on that, but it is still an option.
Any help greatly apprecieated.
Leon

Thanks for your suggestions, but I have managed to work it out.
To capture the image I was using WINGDIAPI HBITMAP WINAPI CreateDIBSection in C++, then OpenGL would draw to that bitmap. Unbeknown to be, there was padding added to the bitmap automatically the width was not a multiple of 4.
Therefore Java was incorrectly interpreting the byte array.
Correct way is to interpret bytes is
byte[] data = captureImage(OpenGLCanvas.getLastFocused().getViewId(), x, y);
int x_padding = x%4;
BufferedImage buff_image = new BufferedImage(x, y, BufferedImage.TYPE_INT_RGB);
int val;
for (int j = 0; j < y; j++)
{
for (int i = 0; i < x; i++)
{
val = ( data[(i + j*x)*3 + j*x_padding + 2]& 0xff) +
((data[(i + j*x)*3 + j*x_padding + 1]& 0xff) << 8) +
((data[(i + j*x)*3 + j*x_padding + 0]& 0xff) << 16);
buff_image.setRGB(i, j, val);
}
}
//save the BufferedImage as a jpeg
try
{
File file = new File(file_name);
FileOutputStream out = new FileOutputStream(file);
JPEGImageEncoder encoder = JPEGCodec.createJPEGEncoder(out);
JPEGEncodeParam param = encoder.getDefaultJPEGEncodeParam(buff_image);
param.setQuality(1.0f, false);
encoder.setJPEGEncodeParam(param);
encoder.encode(buff_image);
out.close();
}

The JPEG standard is extremely complex. I am thinking it may be an issue with padding the output of the DCT somehow. The DCT is done to transform the content from YCrCb 4:2:2 to signal space with one DCT for each channel, Y,Cr, and Cb. The DCT is done on a "Macroblock" or "minimum coded block" depending on your context. JPEG usually has 8x8 macroblocks. When on the edge and there are not enough pixel it clamps the edge value and "drags it across" and does a DCT on that.
I am not sure if this helps, but it sounds like a non standard conforming file. I suggest you use JPEGSnoop to find out more. There are also several explanations about how JPEG compression works.
One possibility is that the sample rate may be encoded incorrectly. It might be something exotic such as 4:2:1 So you might be pulling twice as many X samples as there really are, thus distorting the image.

it is an image I capture from the screen
Maybe the Screen Image class will be easier to use.

Lossless JPEG Rotate (90/180/270 degrees) in Java?

Is there a Java library for rotating JPEG files in increments of 90 degrees, without incurring image degradation?

I found this: http://mediachest.sourceforge.net/mediautil/
API: http://mediachest.sourceforge.net/mediautil/javadocs/mediautil/image/jpeg/LLJTran.html

Building on Henry's answer, here's an example of how to use MediaUtil to perform lossless JPEG rotation based on the EXIF data:
try {
// Read image EXIF data
LLJTran llj = new LLJTran(imageFile);
llj.read(LLJTran.READ_INFO, true);
AbstractImageInfo<?> imageInfo = llj.getImageInfo();
if (!(imageInfo instanceof Exif))
throw new Exception("Image has no EXIF data");
// Determine the orientation
Exif exif = (Exif) imageInfo;
int orientation = 1;
Entry orientationTag = exif.getTagValue(Exif.ORIENTATION, true);
if (orientationTag != null)
orientation = (Integer) orientationTag.getValue(0);
// Determine required transform operation
int operation = 0;
if (orientation > 0
&& orientation < Exif.opToCorrectOrientation.length)
operation = Exif.opToCorrectOrientation[orientation];
if (operation == 0)
throw new Exception("Image orientation is already correct");
OutputStream output = null;
try {
// Transform image
llj.read(LLJTran.READ_ALL, true);
llj.transform(operation, LLJTran.OPT_DEFAULTS
| LLJTran.OPT_XFORM_ORIENTATION);
// Overwrite original file
output = new BufferedOutputStream(new FileOutputStream(imageFile));
llj.save(output, LLJTran.OPT_WRITE_ALL);
} finally {
IOUtils.closeQuietly(output);
llj.freeMemory();
}
} catch (Exception e) {
// Unable to rotate image based on EXIF data
...
}

Regarding the issue of EXIF data not necessarily being handled correctly, since EXIF data is irrelevant in many situations, here's example code demonstrating only the LLJTran lossless JPEG rotation feature (with thanks to user113215):
final File SrcJPEG = new File("my-input.jpg");
final File DestJPEG = new File("my-output.jpg");
final FileInputStream In = new FileInputStream(SrcJPEG);
try {
final LLJTran LLJT = new LLJTran(In);
LLJT.read(LLJTran.READ_ALL, true);
LLJT.transform(LLJTran.ROT_90);
final FileOutputStream Out = new FileOutputStream(DestJPEG);
try {
LLJT.save(Out, LLJTran.OPT_WRITE_ALL);
} finally {
Out.close();
}
} finally {
In.close();
}
If you make the input and output File objects refer to the same file, you can run this over and over again, and observe that the image does not degrade, no matter how many iterations it is put through.

For Android specifically, I found this fork:
https://github.com/juanitobananas/AndroidMediaUtil
Benefits over upstream:
Gradle/Android Studio project
Compatible with jitpack.io
It might even be usable on normal Java, as the code does not import any Android-specific package (I haven't tried though).

You don't need an external library for this kind of thing, it's all built into SE. The easiest being the rotate() function of the Graphics2D object.
For example:
Image rotatedImage = new BufferedImage(imageToRotate.getHeight(null), imageToRotate.getWidth(null), BufferedImage.TYPE_INT_ARGB);
Graphics2D g2d = (Graphics2D) rotatedImage.getGraphics();
g2d.rotate(Math.toRadians(90.0));
g2d.drawImage(imageToRotate, 0, -rotatedImage.getWidth(null), null);
g2d.dispose();
no loss!
Or, if you want to be extra careful, just use BufferedImage.getRGB(x,y), and translate it pixel by pixel on to the new image.

Byte Array to Image object

I am given a byte[] array in Java which contains the bytes for an image, and I need to output it into an image. How would I go about doing this?
Much thanks

BufferedImage img = ImageIO.read(new ByteArrayInputStream(bytes));

If you know the type of image and only want to generate a file, there's no need to get a BufferedImage instance. Just write the bytes to a file with the correct extension.
try (OutputStream out = new BufferedOutputStream(new FileOutputStream(path))) {
out.write(bytes);
}

From Database.
Blob blob = resultSet.getBlob("pictureBlob");
byte [] data = blob.getBytes( 1, ( int ) blob.length() );
BufferedImage img = null;
try {
img = ImageIO.read(new ByteArrayInputStream(data));
} catch (IOException e) {
e.printStackTrace();
}
drawPicture(img); // void drawPicture(Image img);

Since it sounds like you already know what format the byte[] array is in (e.g. RGB, ARGB, BGR etc.) you might be able to use BufferedImage.setRGB(...), or a combination of BufferedImage.getRaster() and WritableRaster.setPixels(...) or WritableRaster.setSamples(...). Unforunately both of these methods require you transform your byte[] into one of int[], float[] or double[] depending on the image format.

According to the Java docs, it looks like you need to use the MemoryImageSource Class to put your byte array into an object in memory, and then use Component.createImage(ImageProducer) next (passing in your MemoryImageSource, which implements ImageProducer).