arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1802.00664cs.CVeess.IV
收藏

基于卷积神经网络回归的数字全息深度预测

Convolutional neural network-based regression for depth prediction in digital holography

Chiba University(千叶大学)

Tomoyoshi Shimobaba, Takashi Kakue, Tomoyoshi Ito

摘要:

数字全息术使我们能够从成像设备捕获的全息图中重建三维空间中的物体。对于重建,我们需要预先知道记录对象的深度位置。在这项研究中,我们提出了使用基于卷积神经网络(CNN)的回归进行深度预测。在以前的研究中,物体的深度是通过使用指示最聚焦的深度位置的一定度量从全息图在不同深度位置处的重构图像来估计的;然而,这样的深度搜索是耗时的。该方法的CNN可以直接从全息图中预测出毫米级精度的深度位置。

英文摘要:

Digital holography enables us to reconstruct objects in three-dimensional space from holograms captured by an imaging device. For the reconstruction, we need to know the depth position of the recoded object in advance. In this study, we propose depth prediction using convolutional neural network (CNN)-based regression. In the previous researches, the depth of an object was estimated through reconstructed images at different depth positions from a hologram using a certain metric that indicates the most focused depth position; however, such a depth search is time-consuming. The CNN of the proposed method can directly predict the depth position with millimeter precision from holograms.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑