Shiyuan YANG and Teruo SHIMOMURA

Department of Electrical Engineering, Faculty of Engineering, Kyushu Institute of Technology, 1-1, Sensui-cho, Tobata-ku, Kitakyushu, 804 Japan

(Received February 12, 1997; Accepted September 29, 1997)

The authors proposed an interlacing technique (IT) approach for the synthesis of kinoform which uses a number of subkinoforms interlaced together into a multikinoform. Using the IT approach, the reconstructed image can be improved by optimizing each subkinoform with the general error-reduction algorithm. In this paper, they show a new iterative algorithm based on the IT, namely an iterative multikinoform method (IMM). After each of the subkinoforms is generated by the IT, the error image, which is the difference between the desired image and the sum of the images reconstructed from all subkinoforms, is circulated into an individual subkinoform to optimize each subkinoform so that the reconstruction error can be reduced further. This process can be performed on all subkinoforms iteratively. Simulation results show that the reconstruction error can be reduced further for kinoforms with larger quantization levels (larger than 2⁴) and its convergence is very rapid when the IMM is used.

Key words : kinoform, iterative algorithm, interlacing technique, subkinoform, multikinoform, reconstruction error

*yang@elcs.kyutech.ac.jp