DCT

Overview

Package

Class

Use

Deprecated

Index

Home >> All >> org >> jfor >> jfor >> tools >> [ jpeg overview ]

PREV CLASS NEXT CLASS

SUMMARY:

JAVADOC |

Sourc code of org.jfor.jfor.tools.jpeg.DCT

SOURCE |

DOWNLOAD | NESTED | FIELD | CONSTR | METHOD

DETAIL: FIELD | CONSTR | METHOD

org.jfor.jfor.tools.jpeg
Class DCT

java.lang.Object
  org.jfor.jfor.tools.jpeg.DCT

class DCT
extends java.lang.Object

DCT - A Java implementation of the Discreet Cosine Transform. This class incorporates quality scaling as implemented in the JPEG-6a library.

Field Summary

private static int dctBlockSize
          DCT Block Size - default 8

private java.lang.Object[] divisors
          Divisors

private java.lang.Object[] quantum
          Quantum

Constructor Summary

(package private) DCT(int quality)
          Constructs a new DCT object.

Method Summary

void dispose()
          Dispose.

(package private) double[][] forwardDCT(float[][] input)
          This method preforms a DCT on a block of image data using the AAN method as implemented in the IJG Jpeg-6a library.

private double[][] forwardDCTExtreme(float[][] input)
          This method preforms forward DCT on a block of image data using the literal method specified for a 2-D Discrete Cosine Transform.

(package private) int[] getQuantumArray(int index)
          Gets a value from quantum.

private void initMatrix(int quality)
          This method sets up the quantization matrix for luminance and chrominance using the Quality parameter.

(package private) int[] quantizeBlock(double[][] inputData, int code)
          This method quantitizes data and rounds it to the nearest integer.

private int[] quantizeBlockExtreme(double[][] inputData, int code)
          This is the method for quantizing a block DCT'ed with forwardDCTExtreme This method quantitizes data and rounds it to the nearest integer.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

quantum

private java.lang.Object[] quantum

Quantum

divisors

private java.lang.Object[] divisors

Divisors

dctBlockSize

private static int dctBlockSize

DCT Block Size - default 8

Constructor Detail

DCT

DCT(int quality)

Constructs a new DCT object. Initializes the cosine transform matrix these are used when computing the DCT and it's inverse. This also initializes the run length counters and the ZigZag sequence. Note that the image quality can be worse than 25 however the image will be extemely pixelated, usually to a block size of N.

Method Detail

dispose

public void dispose()

Dispose.

getQuantumArray

int[] getQuantumArray(int index)

Gets a value from quantum.

quantizeBlock

int[] quantizeBlock(double[][] inputData,
                    int code)

This method quantitizes data and rounds it to the nearest integer.

initMatrix

private void initMatrix(int quality)

This method sets up the quantization matrix for luminance and chrominance using the Quality parameter.

forwardDCTExtreme

private double[][] forwardDCTExtreme(float[][] input)

This method preforms forward DCT on a block of image data using the literal method specified for a 2-D Discrete Cosine Transform. It is included as a curiosity and can give you an idea of the difference in the compression result (the resulting image quality) by comparing its output to the output of the AAN method below. It is ridiculously inefficient. For now the final output is unusable. The associated quantization step needs some tweaking. If you get this part working, please let me know.

forwardDCT

double[][] forwardDCT(float[][] input)

This method preforms a DCT on a block of image data using the AAN method as implemented in the IJG Jpeg-6a library.

quantizeBlockExtreme

private int[] quantizeBlockExtreme(double[][] inputData,
                                   int code)

This is the method for quantizing a block DCT'ed with forwardDCTExtreme This method quantitizes data and rounds it to the nearest integer.