Active nanophotonic devices play an important role in optical communication systems. Improving modulation efficiency and reducing the footprint of these devices are crucial for low energy information processing and on chip communication. The Multi Frequency-Domain Finite Difference (MF-FDFD) algorithm was invented to address the inherent the large difference in scale between the modulation frequency and the frequency of optical waves of active nanophotonic devices. However, due to the fact that the MF-FDFD algorithm itself requires solving a large number of unknowns compared to traditional FDFD algorithms, the current MF-FDFD algorithm still stays in two-dimensional and is difficult to calculate in three-dimensional. To solve this problem, we introduced Newton and Born methods to accelerate and complete the the solution of algorithm, and developed a three-dimensional MF-FDFD algorithm. We validated our algorithm by simulating a mode converter. Our algorithm can effectively perform first principles simulations for active nanophotonic devices, laying the foundation for future intelligent inverse design.
The finite-difference frequency-domain (FDFD) method is an effective method for numerical simulation of electromagnetic fields. It has great advantages in dealing with electromagnetic scattering problems of complex structures and complex media. This method can transform the frequency-domain Maxwell equations into a linear system for solution by difference operation on the spatial grid. However, high-precision differential calculations can result in more memory consumption and a decrease in computational speed. In previous reports, subgridding technique is often used to solve such problems, where mesh refinement is only performed in local areas, while coarse mesh partitioning is still used in other areas. However, the refinement area can only be manually set, lacking flexibility and accuracy. Therefore, we propose a novel FDFD method based on adaptive grids, which uses the cartesian tree-based hierarchical grids to discrete the spatial domain. It can automatically refine the local grids according to the geometrical characteristic of the model to improve the accuracy of specific areas, without significantly increasing the number of unknowns, and has strong flexibility while improving the calculation efficiency. In this study, we use two levels of grids for adaptive grids construction, with a mesh size ratio of 3:1. Using second-order interpolation to handle the transmission problem of electromagnetic field components at different grid boundaries. The simulation results show that the computation speed of the adaptive grids FDFD system is faster than that of structured grids.
The object tracking accuracy may be decreased because of the camera jitter, making it extremely hard for object tracking and trajectory analyzation. To achieve accurate video stabilization, the movement of camera can be analyzed and predicted based on the previous camera jitter sequence. In the area of sequence prediction, the long-short term memory (LSTM) network shows the potential in sequence forecasting, here we use LSTM network in camera jitter prediction and video stabilization. In this paper, we propose a video stabilization algorithm based on multi-region grey projection method and LSTM encoder-decoder network. Our algorithm calculates the motion of the camera through the gray projection of four areas in each frame, then filters out the main movement direction and jitter of the camera. The LSTM encoder-decoder network receives the camera jitter sequence, predicts the camera jitter then stabilizes the video. We to verify the performance of the proposed video stabilization method. We tested the proposed video stabilization algorithm on the jitter videos, which is made by the VisDrone dataset video modified with our recorded camera jitter. Experimental results demonstrate that the proposed method can achieve the video stabilization in real time, and increase the accuracy of object tracking and trajectory analyzation.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.