Grating/microcavity arrays on lily petal replicas as dual-functional SERS substrates for rapid monitoring of thiram on fruit surface and malachite green in surface water.

A dual-functional biomimetic surface-enhanced Raman scattering (SERS) substrate composed of grating/microcavity arrays (GCA) was developed for on-site monitoring of pesticides on fruit peels and in surface water. The hierarchical GCA structure was fabricated by molding a negative polydimethylsiloxane (PDMS) replica from lily petals, followed by the generation of grating wrinkles through oxygen plasma treatment. After the deposition of silver nanoparticles (Ag NPs) through magnetron sputtering, the optimal Ag NPs@GCA substrate exhibits high sensitivity, good uniformity, and reproducibility.

Rapid detection and quantitative analysis of thiram in fruits using a shape-adaptable flexible SERS substrate combined with deep learning.

Ensuring food safety necessitates rapid identification of pesticide residues on fruits. Herein, we developed a shape-adaptable flexible surface-enhanced Raman scattering (SERS) substrate, combined with a deep learning algorithm, to quickly detect and quantitatively analyze thiram on fruit surfaces. This SERS substrate was fabricated by depositing silver nanoparticles (Ag NPs) onto a thin, corrugated polydimethylsiloxane (PDMS) film.

Factorization Vision Transformer: Modeling Long-Range Dependency With Local Window Cost.

Transformers have astounding representational power but typically consume considerable computation which is quadratic with image resolution. The prevailing Swin transformer reduces computational costs through a local window strategy. However, this strategy inevitably causes two drawbacks: 1) the local window-based self-attention (WSA) hinders global dependency modeling capability and 2) recent studies point out that local windows impair robustness.

SCANet: A Unified Semi-Supervised Learning Framework for Vessel Segmentation.

Automatic subcutaneous vessel imaging with near-infrared (NIR) optical apparatus can promote the accuracy of locating blood vessels, thus significantly contributing to clinical venipuncture research. Though deep learning models have achieved remarkable success in medical image segmentation, they still struggle in the subfield of subcutaneous vessel segmentation due to the scarcity and low-quality of annotated data. To relieve it, this work presents a novel semi-supervised learning framework, SCANet, that achieves accurate vessel segmentation through an alternate training strategy.

SiamATL: Online Update of Siamese Tracking Network via Attentional Transfer Learning.

Visual object tracking with semantic deep features has recently attracted much attention in computer vision. Especially, Siamese trackers, which aim to learn a decision making-based similarity evaluation, are widely utilized in the tracking community. However, the online updating of the Siamese fashion is still a tricky issue due to the limitation, which is a tradeoff between model adaption and degradation.