geometric image manipulation.htm

是一部关于java高级图像处理的的一本入门书

        <TD><CODE>samples</CODE> 
          <P></P>
        <TD>A two-dimensional array of ints. 
          <P></P>
      <TR vAlign=top>
        <TD><CODE>xfrac</CODE> 
          <P></P>
        <TD>The <EM>x</EM> subsample position, multiplied by 
          2<SUP>subsampleBits</SUP>. 
          <P></P>
      <TR vAlign=top>
        <TD><CODE>yfrac</CODE> 
          <P></P>
        <TD>The <EM>y</EM> subsample position, multiplied by 
          2<SUP>subsampleBits</SUP>. 
          <P></P></TR></TBODY></TABLE></A>
    <P></P></DD></DL><PRE><UL>
<LI>float interpolate(float[][] samples, float xfrac, float yfrac)
<P></P></LI></UL></PRE>
  <DL><A name=65370>
    <DT>
    <DD>performs interpolation on a two-dimensional array of floating-point 
    samples. This is the same as the above method, only using float values 
    instead of ints. </A>
    <P></P></DD></DL><PRE><UL>
<LI>double interpolate(double[][] samples, float xfrac, 
       float  yfrac)
<P></P></LI></UL></PRE>
  <DL><A name=69291>
    <DT>
    <DD>Performs interpolation on a 2-dimensional array of double samples. </A>
    <P></P></DD></DL><PRE><UL>
<LI>int interpolate(int s00, int s01, int s10, int s11, int xfrac, 
       int yfrac)
<P></P></LI></UL></PRE>
  <DL><A name=65163>
    <DT>
    <DD>performs interpolation on a 2 x 2 grid of integral samples. It should 
    only be called if width == height == 2 and leftPadding == topPadding == 0. 
    </A>
    <P><A name=65243></P>
    <DT>
    <DD>The <CODE>s00</CODE>, <CODE>s01</CODE>, <CODE>s10</CODE>, and 
    <CODE>s11</CODE> parameters are the sample values (see the 2 x 2 grid 
    illustration in <A 
    href="http://java.sun.com/products/java-media/jai/forDevelopers/jai1_0_1guide-unc/Geom-image-manip.doc.html#66155">Figure 
    8-1</A>). </A>
    <P></P></DD></DL><PRE><UL>
<LI>float interpolate(float s00, float s01, float s10, float s11, 
       float xfrac, float yfrac)
<P></P></LI></UL></PRE>
  <DL><A name=65402>
    <DT>
    <DD>performs interpolation on a 2 x 2 grid of integral samples. This is the 
    same as the above method, only using float values instead of ints. </A>
    <P></P></DD></DL><PRE><UL>
<LI>double interpolate(double s00, double s01, double s10, double 
       s11, float xfrac, float yfrac)
<P></P></LI></UL></PRE>
  <DL><A name=69315>
    <DT>
    <DD>performs interpolation on a 2 x 2 grid of double samples. </A>
    <P></P></DD></DL><PRE><UL>
<LI>int interpolate(int s__, int s_0, int s_1, int s_2, int s0_, 
       int  s00, int s01, int s02, int s1_, int s10, int s11, 
       int  s12, int s2_, int s20, int s21, int s22, int xfrac, 
       int  yfrac)
<P></P></LI></UL></PRE>
  <DL><A name=65317>
    <DT>
    <DD>performs interpolation on a 4 x 4 grid of integral samples. It should 
    only be called if width == height == 4 and leftPadding == topPadding == 1. 
    </A>
    <P><A name=65331></P>
    <DT>
    <DD>The <CODE>s__</CODE>, <CODE>through</CODE> <CODE>s22</CODE> parameters 
    are the sample values (see the 4 x 4 grid illustration in <A 
    href="http://java.sun.com/products/java-media/jai/forDevelopers/jai1_0_1guide-unc/Geom-image-manip.doc.html#66155">Figure 
    8-1</A>). </A>
    <P></P></DD></DL><PRE><UL>
<LI>float interpolate(float s__, float s_0, float s_1, float s_2, 
       float s0_, float s00, float s01, float s02, float s1_, 
       float  s10, float s11, float s12, float s2_, float s20, 
       float  s21, float s22, float xfrac, float yfrac)
<P></P></LI></UL></PRE>
  <DL><A name=65477>
    <DT>
    <DD>performs interpolation on a 4 x 4 grid of integral samples. This is the 
    same as the above method, only using float values instead of ints. </A>
    <P></P></DD></DL><PRE><UL>
<LI>abstract int getSubsampleBitsH()
<P></P></LI></UL></PRE>
  <DL><A name=70501>
    <DT>
    <DD>returns the number of bits used to index subsample positions in the 
    horizontal direction. All integral <CODE>xfrac</CODE> parameters should be 
    in the range of 0 to 2<SUP>(getSubsampleBitsH)</SUP> - 1. </A>
    <P></P></DD></DL><PRE><UL>
<LI>int getSubsampleBitsV()
<P></P></LI></UL></PRE>
  <DL><A name=70518>
    <DT>
    <DD>returns the number of bits used to index subsample positions in the 
    vertical direction. All integral <CODE>yfrac</CODE> parameters should be in 
    the range of 0 to 2<SUP>(getSubsampleBitsV)</SUP> - 1. </A>
    <P></P></DD></DL><A name=70504>
  <H3>8.2.1 <IMG 
  src="Geometric Image Manipulation.files/space.gif">Nearest-neighbor 
  Interpolation</H3></A>Nearest-neighbor interpolation, also known as zero-order 
  interpolation, is the fastest interpolation method, though it can produce 
  image artifacts called <EM>jaggies</EM> or <EM>aliasing error</EM>. Jaggies 
  are image artifacts in which the straight edges of objects appear to be rough 
  or jagged. 
  <P>Nearest-neighbor interpolation simply assigns to point <EM>D</EM> in the 
  destination image the value of the pixel nearest <EM>S</EM> in the source 
  image. 
  <P>Neighborhoods of sizes 2 x 1, 1 x 2, 2 x 2, 4 x 1, 1 x 4, 4 x 4, N x 1, and 
  1 x N, that is, all the <CODE>interpolate()</CODE> methods defined in the 
  <CODE>Interpolation</CODE> class, are supported in the interest of simplifying 
  code that handles a number of types of interpolation. In each case, the 
  central sample is returned and the rest are ignored. 
  <P>
  <TABLE border=0>
    <TBODY>
    <TR>
      <TD><IMG src="Geometric Image Manipulation.files/cistine.gif"></TD>
      <TD>
        <HR>
        <B>API:</B> <CODE>javax.media.jai.InterpolationNearest </CODE>
        <HR>
      </TD></TR></TBODY></TABLE><PRE><UL>
<LI>InterpolationNearest()
<P></P></LI></UL></PRE>
  <DL><A name=73430>
    <DT>
    <DD>constructs an <CODE>InterpolationNearest</CODE>. The return value of 
    <CODE>getSubsampleBitsH()</CODE> and <CODE>getSubsampleBitsV()</CODE> will 
    be 0. </A>
    <P></P></DD></DL><A name=55403>
  <H3>8.2.2 <IMG src="Geometric Image Manipulation.files/space.gif">Bilinear 
  Interpolation</H3></A>Bilinear interpolation, also known as first-order 
  interpolation, linearly interpolates pixels along each row of the source 
  image, then interpolates along the columns. Bilinear interpolation assigns to 
  Point <EM>D</EM> in the destination a value that is a bilinear function of the 
  four pixels nearest <EM>S</EM> in the source image. 
  <P>Bilinear interpolation results in an improvement in image quality over 
  nearest-neighbor interpolation, but may still result in less-than-desirable 
  smoothing effects. 
  <P>Bilinear interpolation requires a neighborhood extending one pixel to the 
  right and below the central sample. If the subsample position is given by 
  (<EM>u</EM>, <EM>v</EM>), the resampled pixel value will be: 
  <P>
  <DL><A name=55410>
    <DT>
    <DD><IMG 
    src="Geometric Image Manipulation.files/Geom-image-manip.doc.anc4.gif"> </A>
    <P></P></DD></DL>
  <TABLE border=0>
    <TBODY>
    <TR>
      <TD><IMG src="Geometric Image Manipulation.files/cistine.gif"></TD>
      <TD>
        <HR>
        <B>API:</B> <CODE>javax.media.jai.InterpolationBilinear </CODE>
        <HR>
      </TD></TR></TBODY></TABLE><PRE><UL>
<LI>InterpolationBilinear(int subsampleBits)
<P></P></LI></UL></PRE>
  <DL><A name=55414>
    <DT>
    <DD>constructs an <CODE>InterpolationBilinear</CODE> object with a given 
    subsample precision, in bits.
    <P>
    <TABLE cellPadding=3 border=3>
      <CAPTION><FONT size=-1><B></B></FONT></CAPTION>
      <TBODY>
      <TR vAlign=top>
        <TD><EM>Parameters</EM>: 
          <P></P>
        <TD><CODE>subsampleBits</CODE> 
          <P></P>
        <TD>The subsample precision. 
          <P></P></TR></TBODY></TABLE></A>
    <P></P></DD></DL><PRE><UL>
<LI>InterpolationBilinear()
<P></P></LI></UL></PRE>
  <DL><A name=55423>
    <DT>
    <DD>constructs an <CODE>InterpolationBilinear</CODE> object with the default 
    subsample precision. </A>
    <P></P></DD></DL><A name=55431>
  <H3>8.2.3 <IMG src="Geometric Image Manipulation.files/space.gif">Bicubic 
  Interpolation</H3></A>Bicubic interpolation reduces resampling artifacts even 
  further by using the 16 nearest neighbors in the interpolation and by using 
  bicubic waveforms rather than the linear waveforms used in bilinear 
  interpolation. Bicubic interpolation preserves the fine detail present in the 
  source image at the expense of the additional time it takes to perform the 
  interpolation. 
  <P>The bicubic interpolation routine assigns to point <EM>D</EM> in the 
  destination image a value that is a bicubic function of the 16 pixels nearest 
  <EM>S</EM> in the source image. 
  <P>Bicubic interpolation performs interpolation using the following piecewise 
  cubic polynomial: 
  <P>
  <DL><A name=55445>
    <DT>
    <DD><IMG 
    src="Geometric Image Manipulation.files/Geom-image-manip.doc.anc1.gif"> </A>
    <P></P></DD></DL>Bicubic interpolation requires a neighborhood extending one 
  sample to the left of and above the central sample, and two samples to the 
  right of and below the central sample. 
  <P>
  <TABLE border=0>
    <TBODY>
    <TR>
      <TD><IMG src="Geometric Image Manipulation.files/cistine.gif"></TD>
      <TD>
        <HR>
        <B>API:</B> <CODE>javax.media.jai.InterpolationBicubic </CODE>
        <HR>
      </TD></TR></TBODY></TABLE><PRE><UL>
<LI>InterpolationBicubic(int subsampleBits)
<P></P></LI></UL></PRE>
  <DL><A name=60066>
    <DT>
    <DD>constructs an <CODE>InterpolationBicubic</CODE> with a given subsample 
    precision, in bits.
    <P>
    <TABLE cellPadding=3 border=3>
      <CAPTION><FONT size=-1><B></B></FONT></CAPTION>
      <TBODY>
      <TR vAlign=top>
        <TD><EM>Parameters</EM>: 
          <P></P>
        <TD><CODE>subsampleBits</CODE> 
          <P></P>
        <TD>The subsample precision. 
          <P></P></TR></TBODY></TABLE></A>
    <P></P></DD></DL><A name=55447>
  <H3>8.2.4 <IMG src="Geometric Image Manipulation.files/space.gif">Bicubic2 
  Interpolation</H3></A>Bicubic2 interpolation is basically the same as bicubic 
  interpolation, but uses a different polynomial function. Bicubic2 
  interpolation uses the following piecewise cubic polynomial: 
  <P>
  <DL><A name=55459>
    <DT>
    <DD><IMG 
    src="Geometric Image Manipulation.files/Geom-image-manip.doc.anc3.gif"> </A>
    <P></P></DD></DL>Bicubic interpolation requires a neighborhood extending one 
  sample to the left of and above the central sample, and two samples to the 
  right of and below the central sample. 
  <P>
  <TABLE border=0>
    <TBODY>
    <TR>
      <TD><IMG src="Geometric Image Manipulation.files/cistine.gif"></TD>
      <TD>
        <HR>
        <B>API:</B> <CODE>javax.media.jai.InterpolationBicubic2 </CODE>
        <HR>
      </TD></TR></TBODY></TABLE><PRE><UL>
<LI>InterpolationBicubic2(int subsampleBits)
<P></P></LI></UL></PRE>
  <DL><A name=55480>
    <DT>
    <DD>constructs an <CODE>InterpolationBicubic2</CODE> with a given subsample 
    precision, in bits.
    <P>
    <TABLE cellPadding=3 border=3>
      <CAPTION><FONT size=-1><B></B></FONT></CAPTION>
      <TBODY>
      <TR vAlign=top>