| 灰度视差梯度神经网络与合成全息透视图插值 |
| |
| 引用本文： | 竺子民. 灰度视差梯度神经网络与合成全息透视图插值[J]. 华中科技大学学报(自然科学版), 1999, 27(1): 74-77 |
| |
| 作者姓名： | 竺子民 |
| |
| 作者单位： | 华中理工大学光电子工程系 |
| |
| 摘 要： | 以自然景物三维显示为目标，针对透视图序列现场拍摄数量巨大的困难，提出运用新型视差梯度神经网络压缩透视图数量的构思，给出了拍摄的适定条件以避免遮挡问题．实验表明，该方法匹配准确率高，插值效果好，数据压缩率高
|
| 关 键 词： | 神经网络 视差梯度 合成全息 透视图 |
| 修稿时间： | 1998-06-19 |
Gray Dependent Neural Network of Disparity Gradient and Its Aplication in Interpolation of Perspective Images |
| |
| Zhu Zimin. Gray Dependent Neural Network of Disparity Gradient and Its Aplication in Interpolation of Perspective Images[J]. JOURNAL OF HUAZHONG UNIVERSITY OF SCIENCE AND TECHNOLOGY.NATURE SCIENCE, 1999, 27(1): 74-77 |
| |
| Authors: | Zhu Zimin |
| |
| Abstract: | In order to overcome the difficulty of huge number of photographed perspective images, a new scheme to greatly compress the data with gray dependent network of disparity gradient is presented, and a suitable condition against occlusion for photographing is given. The experiments prove that the scheme features high matching accuracy, good effect of interpolation and high compression rate. |
| |
| Keywords: | neural network disparity gradient synthetic hologram perspective image |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
