colorlookuptable.java

来自「Mac OS X 10.4.9 for x86 Source Code gcc」· Java 代码 · 共 431 行

JAVA
431
字号
/* ColorLookUpTable.java -- ICC v2 CLUT   Copyright (C) 2004 Free Software Foundation   This file is part of GNU Classpath.   GNU Classpath is free software; you can redistribute it and/or modify   it under the terms of the GNU General Public License as published by   the Free Software Foundation; either version 2, or (at your option)   any later version.   GNU Classpath is distributed in the hope that it will be useful, but   WITHOUT ANY WARRANTY; without even the implied warranty of   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU   General Public License for more details.   You should have received a copy of the GNU General Public License   along with GNU Classpath; see the file COPYING.  If not, write to the   Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA   02111-1307 USA.   Linking this library statically or dynamically with other modules is   making a combined work based on this library.  Thus, the terms and   conditions of the GNU General Public License cover the whole   combination.   As a special exception, the copyright holders of this library give you   permission to link this library with independent modules to produce an   executable, regardless of the license terms of these independent   modules, and to copy and distribute the resulting executable under   terms of your choice, provided that you also meet, for each linked   independent module, the terms and conditions of the license of that   module.  An independent module is a module which is not derived from   or based on this library.  If you modify this library, you may extend   this exception to your version of the library, but you are not   obligated to do so.  If you do not wish to do so, delete this   exception statement from your version.*/package gnu.java.awt.color;import java.awt.color.ColorSpace;import java.awt.color.ICC_Profile;import java.nio.ByteBuffer;/** * ColorLookUpTable handles color lookups through a color lookup table, * as defined in the ICC specification. * Both 'mft2' and 'mft1' (8 and 16-bit) type CLUTs are handled. * * This will have to be updated later for ICC 4.0.0 * * @author Sven de Marothy */public class ColorLookUpTable{  /**   * CIE 1931 D50 white point (in Lab coordinates)   */  private static float[] D50 = { 0.96422f, 1.00f, 0.82521f };  /**   * Number of input/output channels   */  int nIn;  /**   * Number of input/output channels   */  int nOut;  int nInTableEntries; // Number of input table entries  int nOutTableEntries; // Number of output table entries  int gridpoints; // Number of gridpoints  int nClut; // This is nOut*(gridpoints**nIn)  double[][] inTable; // 1D input table ([channel][table])  short[][] outTable; // 1D input table ([channel][table])  double[] clut; // The color lookup table  float[][] inMatrix; // input matrix (XYZ only)  boolean useMatrix; // Whether to use the matrix or not.  int[] multiplier;  int[] offsets; // Hypercube offsets   boolean inputLab; // Set if the CLUT input CS is Lab  boolean outputLab; // Set if the CLUT output CS is Lab  /**   * Constructor   * Requires a profile file to get the CLUT from and the tag of the   * CLUT to create. (icSigXToYZTag where X,Y = [A | B], Z = [0,1,2])   */  public ColorLookUpTable(ICC_Profile profile, int tag)  {    useMatrix = false;    switch (tag)      {      case ICC_Profile.icSigAToB0Tag:      case ICC_Profile.icSigAToB1Tag:      case ICC_Profile.icSigAToB2Tag:	if (profile.getColorSpaceType() == ColorSpace.TYPE_XYZ)	  useMatrix = true;	inputLab = false;	outputLab = (profile.getPCSType() == ColorSpace.TYPE_Lab);	break;      case ICC_Profile.icSigBToA0Tag:      case ICC_Profile.icSigBToA1Tag:      case ICC_Profile.icSigBToA2Tag:	if (profile.getPCSType() == ColorSpace.TYPE_XYZ)	  useMatrix = true;	inputLab = (profile.getPCSType() == ColorSpace.TYPE_Lab);	outputLab = false;	break;      default:	throw new IllegalArgumentException("Not a clut-type tag.");      }    byte[] data = profile.getData(tag);    if (data == null)      throw new IllegalArgumentException("Unsuitable profile, does not contain a CLUT.");    // check 'mft'    if (data[0] != 0x6d || data[1] != 0x66 || data[2] != 0x74)      throw new IllegalArgumentException("Unsuitable profile, invalid CLUT data.");    if (data[3] == 0x32)      readClut16(data);    else if (data[3] == 0x31)      readClut8(data);    else      throw new IllegalArgumentException("Unknown/invalid CLUT type.");  }  /**   * Loads a 16-bit CLUT into our data structures   */  private void readClut16(byte[] data)  {    ByteBuffer buf = ByteBuffer.wrap(data);    nIn = data[8] & (0xFF);    nOut = data[9] & (0xFF);    nInTableEntries = buf.getShort(48);    nOutTableEntries = buf.getShort(50);    gridpoints = data[10] & (0xFF);    inMatrix = new float[3][3];    for (int i = 0; i < 3; i++)      for (int j = 0; j < 3; j++)	inMatrix[i][j] = ((float) (buf.getInt(12 + (i * 3 + j) * 4))) / 65536.0f;    inTable = new double[nIn][nInTableEntries];    for (int channel = 0; channel < nIn; channel++)      for (int i = 0; i < nInTableEntries; i++)	inTable[channel][i] = (double) ((int) buf.getShort(52	                                                   + (channel * nInTableEntries	                                                   + i) * 2)	                      & (0xFFFF)) / 65536.0;    nClut = nOut;    multiplier = new int[nIn];    multiplier[nIn - 1] = nOut;    for (int i = 0; i < nIn; i++)      {	nClut *= gridpoints;	if (i > 0)	  multiplier[nIn - i - 1] = multiplier[nIn - i] * gridpoints;      }    int clutOffset = 52 + nIn * nInTableEntries * 2;    clut = new double[nClut];    for (int i = 0; i < nClut; i++)      clut[i] = (double) ((int) buf.getShort(clutOffset + i * 2) & (0xFFFF)) / 65536.0;    outTable = new short[nOut][nOutTableEntries];    for (int channel = 0; channel < nOut; channel++)      for (int i = 0; i < nOutTableEntries; i++)	outTable[channel][i] = buf.getShort(clutOffset	                                    + (nClut	                                    + channel * nOutTableEntries + i) * 2);    // calculate the hypercube corner offsets    offsets = new int[(1 << nIn)];    offsets[0] = 0;    for (int j = 0; j < nIn; j++)      {	int factor = 1 << j;	for (int i = 0; i < factor; i++)	  offsets[factor + i] = offsets[i] + multiplier[j];      }  }  /**   * Loads a 8-bit CLUT into our data structures.   */  private void readClut8(byte[] data)  {    ByteBuffer buf = ByteBuffer.wrap(data);    nIn = (data[8] & (0xFF));    nOut = (data[9] & (0xFF));    nInTableEntries = 256; // always 256    nOutTableEntries = 256; // always 256    gridpoints = (data[10] & (0xFF));    inMatrix = new float[3][3];    for (int i = 0; i < 3; i++)      for (int j = 0; j < 3; j++)	inMatrix[i][j] = ((float) (buf.getInt(12 + (i * 3 + j) * 4))) / 65536.0f;    inTable = new double[nIn][nInTableEntries];    for (int channel = 0; channel < nIn; channel++)      for (int i = 0; i < nInTableEntries; i++)	inTable[channel][i] = (double) ((int) buf.get(48	                                              + (channel * nInTableEntries	                                              + i)) & (0xFF)) / 255.0;    nClut = nOut;    multiplier = new int[nIn];    multiplier[nIn - 1] = nOut;    for (int i = 0; i < nIn; i++)      {	nClut *= gridpoints;	if (i > 0)	  multiplier[nIn - i - 1] = multiplier[nIn - i] * gridpoints;      }    int clutOffset = 48 + nIn * nInTableEntries;    clut = new double[nClut];    for (int i = 0; i < nClut; i++)      clut[i] = (double) ((int) buf.get(clutOffset + i) & (0xFF)) / 255.0;    outTable = new short[nOut][nOutTableEntries];    for (int channel = 0; channel < nOut; channel++)      for (int i = 0; i < nOutTableEntries; i++)	outTable[channel][i] = (short) (buf.get(clutOffset + nClut	                                        + channel * nOutTableEntries	                                        + i) * 257);    // calculate the hypercube corner offsets    offsets = new int[(1 << nIn)];    offsets[0] = 0;    for (int j = 0; j < nIn; j++)      {	int factor = 1 << j;	for (int i = 0; i < factor; i++)	  offsets[factor + i] = offsets[i] + multiplier[j];      }  }  /**   * Performs a lookup through the Color LookUp Table.   * If the CLUT tag type is AtoB the conversion will be from the device   * color space to the PCS, BtoA type goes in the opposite direction.   *   * For convenience, the PCS values for input or output will always be   * CIE XYZ (D50), if the actual PCS is Lab, the values will be converted.   *   * N-dimensional linear interpolation is used.   */  float[] lookup(float[] in)  {    float[] in2 = new float[in.length];    if (useMatrix)      {	for (int i = 0; i < 3; i++)	  in2[i] = in[0] * inMatrix[i][0] + in[1] * inMatrix[i][1]	           + in[2] * inMatrix[i][2];      }    else if (inputLab)      in2 = XYZtoLab(in);    else      System.arraycopy(in, 0, in2, 0, in.length);    // input table     for (int i = 0; i < nIn; i++)      {	int index = (int) Math.floor(in2[i] * (double) (nInTableEntries - 1)); // floor in	// clip values.	if (index >= nInTableEntries - 1)	  in2[i] = (float) inTable[i][nInTableEntries - 1];	else if (index < 0)	  in2[i] = (float) inTable[i][0];	else	  {	    // linear interpolation	    double alpha = in2[i] * ((double) nInTableEntries - 1.0) - index;	    in2[i] = (float) (inTable[i][index] * (1 - alpha)	             + inTable[i][index + 1] * alpha);	  }      }    // CLUT lookup    double[] output2 = new double[nOut];    double[] weights = new double[(1 << nIn)];    double[] clutalpha = new double[nIn]; // interpolation values    int offset = 0; // = gp    for (int i = 0; i < nIn; i++)      {	int index = (int) Math.floor(in2[i] * ((double) gridpoints - 1.0));	double alpha = in2[i] * ((double) gridpoints - 1.0) - (double) index;	// clip values.	if (index >= gridpoints - 1)	  {	    index = gridpoints - 1;	    alpha = 1.0;	  }	else if (index < 0)	  index = 0;	clutalpha[i] = alpha;	offset += index * multiplier[i];      }    // Calculate interpolation weights    weights[0] = 1.0;    for (int j = 0; j < nIn; j++)      {	int factor = 1 << j;	for (int i = 0; i < factor; i++)	  {	    weights[factor + i] = weights[i] * clutalpha[j];	    weights[i] *= (1.0 - clutalpha[j]);	  }      }    for (int i = 0; i < nOut; i++)      output2[i] = weights[0] * clut[offset + i];    for (int i = 1; i < (1 << nIn); i++)      {	int offset2 = offset + offsets[i];	for (int f = 0; f < nOut; f++)	  output2[f] += weights[i] * clut[offset2 + f];      }    // output table     float[] output = new float[nOut];    for (int i = 0; i < nOut; i++)      {	int index = (int) Math.floor(output2[i] * ((double) nOutTableEntries	                             - 1.0));	// clip values.	if (index >= nOutTableEntries - 1)	  output[i] = outTable[i][nOutTableEntries - 1];	else if (index < 0)	  output[i] = outTable[i][0];	else	  {	    // linear interpolation	    double a = output2[i] * ((double) nOutTableEntries - 1.0)	               - (double) index;	    output[i] = (float) ((double) ((int) outTable[i][index] & (0xFFFF)) * (1	                - a)	                + (double) ((int) outTable[i][index + 1] & (0xFFFF)) * a) / 65536f;	  }      }    if (outputLab)      return LabtoXYZ(output);    return output;  }  /**   * Converts CIE Lab coordinates to (D50) XYZ ones.   */  private float[] LabtoXYZ(float[] in)  {    // Convert from byte-packed format to a     // more convenient one (actual Lab values)    // (See ICC spec for details)    // factor is 100 * 65536 / 65280    in[0] = (float) (100.392156862745 * in[0]);    in[1] = (in[1] * 256.0f) - 128.0f;    in[2] = (in[2] * 256.0f) - 128.0f;    float[] out = new float[3];    out[1] = (in[0] + 16.0f) / 116.0f;    out[0] = in[1] / 500.0f + out[1];    out[2] = out[1] - in[2] / 200.0f;    for (int i = 0; i < 3; i++)      {	double exp = out[i] * out[i] * out[i];	if (exp <= 0.008856)	  out[i] = (out[i] - 16.0f / 116.0f) / 7.787f;	else	  out[i] = (float) exp;	out[i] = D50[i] * out[i];      }    return out;  }  /**   * Converts CIE XYZ coordinates to Lab ones.   */  private float[] XYZtoLab(float[] in)  {    float[] temp = new float[3];    for (int i = 0; i < 3; i++)      {	temp[i] = in[i] / D50[i];	if (temp[i] <= 0.008856f)	  temp[i] = (7.7870689f * temp[i]) + (16f / 116.0f);	else	  temp[i] = (float) Math.exp((1.0 / 3.0) * Math.log(temp[i]));      }    float[] out = new float[3];    out[0] = (116.0f * temp[1]) - 16f;    out[1] = 500.0f * (temp[0] - temp[1]);    out[2] = 200.0f * (temp[1] - temp[2]);    // Normalize to packed format    out[0] = (float) (out[0] / 100.392156862745);    out[1] = (out[1] + 128f) / 256f;    out[2] = (out[2] + 128f) / 256f;    for (int i = 0; i < 3; i++)      {	if (out[i] < 0f)	  out[i] = 0f;	if (out[i] > 1f)	  out[i] = 1f;      }    return out;  }}

⌨️ 快捷键说明

复制代码Ctrl + C
搜索代码Ctrl + F
全屏模式F11
增大字号Ctrl + =
减小字号Ctrl + -
显示快捷键?