Digital maps of road networks are a vital part of digital cities and intelligent transportation. In this paper, we provide a comprehensive review on road extraction based on various remote sensing data sources, including high-resolution images, hyperspectral images, synthetic aperture radar images, and light detection and ranging. This review is divided into three parts. Part 1 provides an overview of the existing data acquisition techniques for road extraction, including data acquisition methods, typical sensors, application status, and prospects. Part 2 underlines the main road extraction methods based on four data sources. In this section, road extraction methods based on different data sources are described and analysed in detail. Part 3 presents the combined application of multisource data for road extraction. Evidently, different data acquisition techniques have unique advantages, and the combination of multiple sources can improve the accuracy of road extraction. The main aim of this review is to provide a comprehensive reference for research on existing road extraction technologies.

Femtosecond (fs)-laser hyperdoped silicon has aroused great interest for applications in infrared photodetectors due to its special properties. Crystallinity and optical absorption influenced by co-hyperdoped nitrogen in surface microstructured silicon, prepared by fs-laser irradiation in gas mixture of SF₆/NF₃ and SF₆/N₂ were investigated. In both gas mixtures, nitrogen and sulfur were incorporated at average concentrations above 10¹⁹ atoms/cm³ in the 20–400 nm surface layer. Different crystallinity and optical absorption properties were observed for samples microstructured in the two gas mixtures. For samples prepared in SF₆/N₂, crystallinity and light absorption properties were similar to samples formed in SF₆. Significant differences were observed amongst samples formed in SF₆/NF₃, which possess higher crystallinity and strong sub-band gap absorption. The differing crystallinity and light absorption rates between the two types of nitrogen co-hyperdoped silicon were attributed to different nitrogen configurations in the doped layer. This was induced by fs-laser irradiating silicon in the two N-containing gas mixtures.

Femtosecond-laser-fabricated black silicon has been widely used in the fields of solar cells, photodetectors, semiconductor devices, optical coatings, and quantum computing. However, the responsive spectral range limits its application in the near- to mid-infrared wavelengths. To further increase the optical responsivity in longer wavelengths, in this work, silicon (Si) was co-hyperdoped with nitrogen (N) and selenium (Se) through the deposition of Se films on Si followed by femtosecond (fs)-laser irradiation in an atmosphere of NF₃. The optical and crystalline properties of the Si:N/Se were found to be influenced by the precursor Se film and laser fluence. The resulting photodetector, a product of this innovative approach, exhibited an impressive responsivity of 24.8 A/W at 840 nm and 19.8 A/W at 1060 nm, surpassing photodetectors made from Si:N, Si:S, and Si:S/Se (the latter two fabricated in SF6). These findings underscore the co-hyperdoping method’s potential in significantly improving optoelectronic device performance.

To address the challenge of balancing multi-scale object recognition accuracy and computational efficiency for real-time detection tasks on unmanned aerial vehicle(UAV) edge computing platforms, we propose an Edge-Attention Synergy Network(EASNet). First, a spatial pyramid information architecture is constructed to enable collaboration perception of local texture features and global semantic information through cross-layer feature interaction mechanism. Second, a channel-adaptive dynamic weighting module is designed, leveraging a differential gating strategy to resolve semantic conflicts in multi-resolution feature fusion. Finally, we propose a hybrid metric loss with learnable normalization. It dynamically adjusts Wasserstein scale sensitivity and incorporates fixed-weight IoU constraints, boosting small-target localization in complex backgrounds. Experimental results demonstrate that our method achieves mAP₅₀ scores of 95.6%, 93.4%, 78.9%, 58.5% and 46.5% on the RSOD, NWPU VHR-10, DIOR, VEDAI and VisDrone datasets, respectively, with improvements ranging from 4.2% to 8.3% over the YOLOv10s baseline. Furthermore, through embeded deployment validation, we successfully implemented the model on a Jetson Nano edge computing platform, attaining 34 FPS for real-time vehicle and pedestrian detection in campus monitoring scenarios.