Thin film characterization by learning-assisted multi-angle polarized microscopy.

Thin film characterization is a necessary step in the semiconductor industry and nanodevice fabrication. In this work, we report a learning-assisted method to conduct the measurement based on a multi-angle polarized microscopy. By illuminating the film with a tightly focused vectorial beam with space-polarization nonseparability, the angle-dependent reflection coefficients are encoded into the reflected intensity distribution. The measurement is then transformed into an optimization problem aiming at minimizing the discrepancy between measured and simulated image features. The proposed approach is validated by numerical simulation and experimental measurements. As the method can be easily implemented with a conventional microscope, it provides a low cost solution to measure film parameters with a high spatial resolution and time efficiency.

OBMO: One Bounding Box Multiple Objects for Monocular 3D Object Detection.

Compared to typical multi-sensor systems, monocular 3D object detection has attracted much attention due to its simple configuration. However, there is still a significant gap between LiDAR-based and monocular-based methods. In this paper, we find that the ill-posed nature of monocular imagery can lead to depth ambiguity. Specifically, objects with different depths can appear with the same bounding boxes and similar visual features in the 2D image. Unfortunately, the network cannot accurately distinguish different depths from such non-discriminative visual features, resulting in unstable depth training. To facilitate depth learning, we propose a simple yet effective plug-and-play module, One Bounding Box Multiple Objects (OBMO). Concretely, we add a set of suitable pseudo labels by shifting the 3D bounding box along the viewing frustum. To constrain the pseudo-3D labels to be reasonable, we carefully design two label scoring strategies to represent their quality. In contrast to the original hard depth labels, such soft pseudo labels with quality scores allow the network to learn a reasonable depth range, boosting training stability and thus improving final performance. Extensive experiments on KITTI and Waymo benchmarks show that our method significantly improves state-of-the-art monocular 3D detectors by a significant margin (The improvements under the moderate setting on KITTI validation set are 1.82 ~ 10.91% mAP in BEV and 1.18 ~ 9.36% mAP in 3D). Codes have been released at https://github.com/mrsempress/OBMO.

Thermally Conductive and Antistatic Properties of Silicone Rubber Reinforced by the Modified Graphene Oxide.

Silicone rubber (SR)/vinyl-graphene oxide (vinyl-GO) nanocomposites were prepared through the hydrosilylation reaction of silicon hydrogen polydimethylsiloxane (H-PDMS) with vinyl polydimethylsiloxane (vinyl-PDMS), in which vinyl-GO was used as a nano filler. The thermally conductive and antistatic properties of the nanocomposites, and their tensile strength and thermal stability were evaluated. The thermally conductive and antistatic properties increased naturally when the nanocomposites had eight to nine parts of vinyl-GO. The addition of 9 parts of vinyl-GO increased the thermal conductivity to 0.44 from 0.17 W/m-1·K-1 of neat SR and the surface resistance value to 108 from 1014 Ω of neat SR. Vinyl-GO is effective in improving the tensile strength and toughness of the nanocomposites. The tensile strength and elongation at break of the nanocomposites were much higher than that of neat SR, especially for 10 parts of vinyl-GO in the nanocomposite, and the tensile strength was 1.84 MPa and the elongation at break was 314.1%. Additionally, compared with neat SR, the nanocomposites had a much higher thermal stability. For eight parts of vinyl-GO in the nanocomposites, H-PDMS with the selected silicon hydrogen content and vinyl-PDMS with the selected vinyl content could offer an appropriate cross-linking degree that suits the character of GO. When the nanocomposite had eight parts of vinyl-GO, its scanning electron microscope exhibited a monolayer GO with folded, twisted, and local surface folds. However, there was a certain amount of multilayer aggregation of GO for 10 parts of vinyl-GO in the nanocomposite.

Phytoplankton Composition and Their Related Factors in Five Different Lakes in China: Implications for Lake Management.

In this paper, two trophic lakes: Lake Taihu and Lake Yanghe, and three alpine lakes: Lake Qinghai, Lake Keluke, and Lake Tuosu, were investigated to discover the connections between environmental factors and the phytoplankton community in lakes with differences in trophic levels and climatic conditions. Three seasonal data, including water quality and phytoplankton, were collected from the five lakes. The results demonstrated clear differences in water parameters and phytoplankton compositions in different lakes. The phytoplankton was dominated by Bacillariophyta, followed by Cyanobacteria and Chlorophyta in Lake Qinghai, Lake Keluke, and Lake Tuosu. It was dominated by Cyanobacteria (followed by Chlorophyta and Bacillariophyta in Lake Yanghe) and Cyanobacteria (followed by Chlorophyta and Cryptophyta in Lake Taihu). The temperature was an essential factor favoring the growth of Cyanobacteria, Chlorophyta, and Bacillariophyta, especially Cyanobacteria and Chlorophyta. The pH had significantly negative relationships with Cyanobacteria, Chlorophyta, and Bacillariophyta. Particularly, a high pH might be a strong and negative factor for phytoplankton growth in alpine lakes. A high salinity was also an adverse factor for phytoplankton. Those results could provide fundamental information about the phytoplankton community and their correlated factors in the alpine lakes of the Tibetan Plateau, contributing to the protection and management of alpine lakes.