Yang S, Shimomura T
Appl Opt. 1996 Dec 10;35(35):6983-9. doi: 10.1364/AO.35.006983.
An interlacing technique algorithm is proposed for the synthesis of kinoforms. The conventional iterative methods are quite powerful for optimizing kinoforms, but there is still a large reconstruction error for a quantized kinoform. We suggest the use of a number of subkinoforms interlaced together to synthesize a multikinoform for reconstructing the desired image. The idea of our interlacing technique is to increase the size of a kinoform to reduce the reconstruction error. The first subkinoform is generated from the desired image. Other subkinoforms are generated from the error images between the desired image and the image reconstructed from the previous subkinoforms. A theoretical analysis shows that the reconstruction error will be reduced as the number of subkinoforms is increased. Simulation results show that our interlacing method can reduce the reconstruction error more than do the conventional iterative methods and that the reconstructed image can be improved.
提出了一种用于合成相息图的交织技术算法。传统的迭代方法在优化相息图方面非常强大,但对于量化的相息图仍存在较大的重建误差。我们建议使用多个交织在一起的子相息图来合成一个多相息图,以重建所需图像。我们交织技术的思想是增大相息图的尺寸以减小重建误差。第一个子相息图由所需图像生成。其他子相息图由所需图像与从前一个子相息图重建的图像之间的误差图像生成。理论分析表明,随着子相息图数量的增加,重建误差将减小。仿真结果表明,我们的交织方法比传统迭代方法能更有效地减小重建误差,并且重建图像质量可以得到提高。