We propose an optical watermarking strategy based on precise bias tuning of an electro-optical modulator and a convolutional neural network-enabled decoder. This scheme can be used for identity aut…
800G Ethernet is expected to be the dominant solution for the next generation inter- and intra-data-center connections. To boost the transmission capacity to 800 Gb/s, utilizing multi-level pulse a…
Continuous casting in steel production uses specially developed oxyfluoride glasses (mold fluxes) to lubricate a mold and control the solidification of the steel in the mold. The composition of the…
Photonic-crystal surface-emitting laser (PCSEL) holds promise for laser beam with high efficacy while retain the small divergence angle, being now all the rage as laser source in modern opto-electr…
A compact fiber optic hydrogen sensing system employing self-referenced configuration and controllable light heating technologies, is proposed and experimentally demonstrated with ultra-high sensit…
GaN-based light emitting diodes (LEDs) have potential as photodetectors (PDs) for visible light communication (VLC). In current studies on LED-based PDs, devices are operated in a reverse-biased or…
Highly sensitive reflective-type fiber hydrogen sensor based on an enlarged taper cascaded with a tilted fiber grating (TFBG) was proposed. Hydrogen-sensitive Au-Pd nanofilm was coated on the TFBG …
Conventional deformable registration methods aim at solving an optimization model carefully designed on image pairs and their computational costs are exceptionally high. In contrast, recent deep le…
Dynamic neural network is an emerging research topic in deep learning. Compared to static models which have fixed computational graphs and parameters at the inference stage, dynamic networks can ad…
Point cloud instance segmentation has achieved huge progress with the emergence of deep learning. However, these methods are usually data-hungry with expensive and time-consuming dense point cloud …