Interlacing technique approach for the synthesis of kinoforms

Error dependence of quantized kinoform reconstruction on the position of the desired image

Two-dimensional Fourier transform kinoforms can be calculated by use of a discrete Fourier transform. It is well known that the off-axis reconstruction has lower reconstruction error than the on-axis one. Here we make what to our knowledge is a new analysis on the effect of phase quantization in the Fourier domain. We find that the kinoform reconstruction error changes periodically according to the position of the desired image when a large dummy area is added. The error dependence of quantized kinoform reconstruction is simulated on the position of the desired image by use of the iterative dummy area method and the iterative interlacing technique.

Error reduction of quantized kinoforms by means of increasing the kinoform size

There are two kinds of method that utilize the redundancy in kinoform domains for reducing the reconstruction errors of quantized kinoforms. One is the iterative-dummy area (IDA) method, which increases the kinoform size indirectly by the addition of a dummy area to the desired image. The other is the interlacing technique (IT), which increases the kinoform size directly by the interlacing of a number of subkinoforms whose sizes are the same as the desired image. We compare the error reduction of quantized kinoforms between these two methods. Simulation results show that reconstruction errors from the IT method can be reduced further and faster than those from the IDA method when the kinoform size is increased to larger than 4 x 4 times the size of the desired image.