Diffractive deep neural networks (DNNs) typically adopt a densely cascaded arrangement of diffractive masks, leading to multiple reflections of diffracted light between adjacent masks, thereby affecting the network's inference capability. It is challenging to fully simulate this multiple-reflection phenomenon. To eliminate this phenomenon, we designed tilted-mode all-optical diffractive deep neural networks (T-DNNs) and proposed a theoretical model for diffraction propagation in the tilted mode. Simulation results indicate that T-DNNs address the performance degradation caused by interlayer reflections in DNNs constructed with high-index diffractive masks. In classification tasks, T-DNNs achieve better classification results compared to DNNs that consider interlayer reflections.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.3390/mi16010008 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
High-temperature structural disorders stabilize hydrous aluminosilicates in the mantle transition zone.
Nat Commun
January 2025
Center for High Pressure Science, State Key Laboratory of Metastable Materials Science and Technology, Yanshan University, Qinhuangdao, 066004, China.
Hydrous aluminosilicates are important deep water-carriers in sediments subducting into the deep mantle. To date, it remains enigmatic how hydrous aluminosilicates withstand extremely high temperatures in the mantle transition zone. Here we systematically investigate the crystal structures and chemical compositions of typical hydrous aluminosilicates using single-crystal X-ray diffraction, electron probe microanalyzer, and nanoscale secondary ion mass spectrometry.
View Article and Find Full Text PDFTilted-Mode All-Optical Diffractive Deep Neural Networks.
Micromachines (Basel)
December 2024
Liaoning Key Laboratory of Marine Sensing and Intelligent Detection, Dalian Maritime University, Dalian 116026, China.
Diffractive deep neural networks (DNNs) typically adopt a densely cascaded arrangement of diffractive masks, leading to multiple reflections of diffracted light between adjacent masks, thereby affecting the network's inference capability. It is challenging to fully simulate this multiple-reflection phenomenon. To eliminate this phenomenon, we designed tilted-mode all-optical diffractive deep neural networks (T-DNNs) and proposed a theoretical model for diffraction propagation in the tilted mode.
View Article and Find Full Text PDFNanocellulose prepared from shiitake mushroom (Lentinus edodes) stipe by high pressure homogenization and the gel-like emulsions stabilized by them.
Int J Biol Macromol
January 2025
National Engineering Research Center for Seafood, State Key Laboratory of Marine Food Processing and Safety Control, Collaborative Innovation Center of Provincial and Ministerial Co-construction for Seafood Deep Processing, Liaoning Province Collaborative Innovation Center for Marine Food Deep Processing, Dalian Technology Innovation Center for Chinese Pre-made Food, College of Food Science and Technology, Dalian Polytechnic University, Dalian 116034, China. Electronic address:
Shiitake mushroom is popularly consumed thanks to its umami taste and good flavor, but its stipe is often discarded due to the rough texture and poor chewiness. In the study, high-pressure homogenization (HPH) was applied to modify the physiochemical properties of shiitake mushroom nanocellulose (SMNC), and the SMNCs were used to constructing gel-like emulsions (EGs). Atomic force microscope and cryo-scanning electron microscope observations showed that SMNCs had shorter length after HPH treatment.
View Article and Find Full Text PDFHalogenated meroterpenoids with antifungal activities from the Deep-Sea-Derived fungus Acremonium sclerotigenum guided by the genomic and OSMAC strategy.
Bioorg Chem
January 2025
State Key Laboratory of Mycology, Institute of Microbiology, Chinese Academy of Sciences, Beijing 100101, China; University of Chinese Academy of Sciences, Beijing 100049, China. Electronic address:
Thirteen new meroterpenoids, acremorins A-M (1, 2, 4, 6, 7 and 9-16), together with three known analogues (3, 5 and 8) were isolated from the deep-sea-derived fungus Acremonium sclerotigenum LW14 guided by the genomic and OSMAC strategy. Their structures and absolute configurations were established by extensive spectroscopic analysis, electronic circular dichroism (ECD) calculations, Rh(OCOCF)-induced ECD experiments, and a single-crystal X-ray diffraction experiment. Compounds 2, 4, 6 and 9 represent the rare brominated ascochlorins.
View Article and Find Full Text PDFNanophotonic inspection of deep-subwavelength integrated optoelectronic chips.
Sci Adv
January 2025
Guangdong Provincial Key Laboratory of Optical Fiber Sensing and Communications, Institute of Photonics Technology, Jinan University, Guangzhou 510632, China.
Artificial nanostructures with ultrafine and deep-subwavelength features have emerged as a paradigm-shifting platform to advanced light-field management, becoming key building blocks for high-performance integrated optoelectronics and flat optics. However, direct optical inspection of integrated chips remains a missing metrology gap that hinders quick feedback between design and fabrications. Here, we demonstrate that photothermal nonlinear scattering microscopy can be used for direct imaging and resolving of integrated optoelectronic chips beyond the diffraction limit.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!