com.c3d.image.hvq
Class HVQtrainer
java.lang.Object
|
+--com.c3d.image.hvq.HVQtrainer
- public class HVQtrainer
- extends java.lang.Object
|
Constructor Summary |
HVQtrainer(int xbitsin,
int ybitsin,
int bitsout)
TeX Intialise with the number of bits precision of the
x and y components of the vectors and the number
of bits precision of the output table. |
| Methods inherited from class java.lang.Object |
clone,
equals,
finalize,
getClass,
hashCode,
notify,
notifyAll,
toString,
wait,
wait,
wait |
HVQtrainer
public HVQtrainer(int xbitsin,
int ybitsin,
int bitsout)
- TeX Intialise with the number of bits precision of the
x and y components of the vectors and the number
of bits precision of the output table.
This should be $ log_2 N$ where N is the number
of codebook entries.
Currently set up to use the supercede persistent store
to hold state of training. Assumes that the trainer
has been allocated space in persistent memory
takeSample
public void takeSample(int x,
int y)
- Give a single data point to the system
getQuantizer
public int[][] getQuantizer()
getCodebook
public int[][] getCodebook()