Object detection in low-light environments has been widely recognized as a critical research direction in the field of computer vision. In response to the challenges of reduced detection accuracy and high edge-deployment costs encountered by mainstream single-stage object detection models under low-light conditions, this paper proposes a lightweight object detection network based on YOLOv11, integrates StarNet, C3k2-Star, and a lightweight detail-enhanced convolution and shared convolutional detection head(LSDECD), so called SCL-YOLOv11 herein. First, the StarNet architecture is introduced into the Backbone to enhance the extraction of shallow image features and significantly reduce computational complexity. Next, Star Blocks from the StarNet framework are employed to optimize the C3k2 module in the Neck stage, thereby improving the localization accuracy of deep features without increasing network complexity. Meanwhile, the Minimum Point Distance based IoU(MPDIoU) loss function is adopted to mitigate gradient explosion risks while enhancing detection precision. Furthermore, a lightweight detail-enhancement convolution layer and a shared-convolution detection head are designed to improve the model’s capability in capturing fine-grained details. Finally, a knowledge distillation procedure is conducted using SCL-YOLOv11 as the student model and YOLOv8 as the teacher model, ensuring that the proposed SCL-YOLOv11 model achieves both low complexity and excellent detection performance. Experimental results on the ExDark dataset demonstrate that the proposed algorithm achieves 67.6% mAP@0.5 and 42.4% mAP@0.5:0.95, while reducing parameter count by 38.5% and computational cost by 25.4%. Consequently, the proposed model is highly suitable for edge deployment in industrial applications and embedded development systems.

Prestressed concrete (PC) continuous beam bridges are widely used in transportation infrastructure. However, their construction involves substantial material consumption, raising sustainability concerns amid increasing environmental pressures. This study aims to address the urgent need for resource-efficient bridge design by developing a comprehensive optimization framework that minimizes material usage while ensuring structural safety, durability, and compliance with engineering standards. The proposed methodology integrates a Genetic Algorithm (GA) with a Backpropagation (BP) neural network to optimize both the cross-sectional geometry and the overall alignment of PC continuous beam bridges. The GA is utilized to identify optimal cross-sectional parameters within regulatory constraints, while the BP neural network, trained on extensive design data, refines the bridge bottom height profile to enhance structural performance. The integrated GA-BP framework is validated through a case study of a continuous beam bridge, demonstrating a 94% improvement in design efficiency, a 14% reduction in concrete consumption, and a 34% reduction in prestressed steel usage during the preliminary design stage. These results highlight the framework’s significant potential in advancing sustainable and intelligent bridge design, offering a novel approach to combining artificial intelligence with structural optimization for practical engineering applications.

Intelligent analysis of 3D point clouds has become a frontier in emerging fields such as autonomous driving, digital twins, and the metaverse. Precise segmentation of 3D point clouds is particularly important within these domains; however, it faces several challenges: (1) point cloud data inherently lacks structured topological information; (2) point cloud shapes are complex and highly variable, making it difficult to utilize semantic priors; and (3) the sampling process of point clouds may result in sparse and uneven data. To address these issues, this paper proposes a novel Point Cloud Segmentation Network based on multi-scale feature fusion and Transformer architecture (MFFTNet). MFFTNet enhances the performance of existing segmentation methods by globally modeling the overall point cloud shape and embedding local point cloud details. Specifically, MFFTNet divides the segmentation task into encoding and decoding stages. The encoder is designed as a hierarchical pyramid structure that extracts relatively sparse local center points and fuses local features during progressive downsampling. It also utilizes a Transformer for global feature modeling to establish multi-scale topological and semantic information of the point cloud. Subsequently, multi-scale feature fusion further enhances the network’s perception of local features and global structure. The decoder progressively upsamples to restore the original point cloud and injects multi-scale feature information to achieve precise segmentation. Based on the aforementioned encoding-decoding structure and multi-scale feature fusion, MFFTNet outperforms existing methods on the point cloud semantic segmentation datasets ShapeNetPart and S3DIS.

In the context of digital construction, responsibility management in smart city building information modeling (BIM) projects spans the entire building lifecycle. The involvement of numerous BIM designers in project management and frequent data exchanges pose significant challenges for the traceability, immutability, and responsibility attribution of BIM models. To address these issues, this study proposes a blockchain-based responsibility management and collaboration framework for BIM projects using non-fungible tokens (NFTs), aiming to enhance the management of responsibilities and accountability in BIM projects. This research adopts a design science methodology, strictly adhering to scientific research procedures to ensure rigor. First, NFTs based on blockchain technology were developed to generate corresponding digital signatures for BIM model files. This approach ensures that each BIM model file has a unique digital identity, enhancing transparency and traceability in responsibility management. Next, the interplanetary file system (IPFS) was used to generate digital fingerprints, with the content identifier generated by IPFS uploaded to the blockchain to ensure the immutability of BIM model files. This method guarantees the integrity and security of BIM model files throughout their lifecycle. Finally, the proposed methods were validated through a blockchain network. The experimental results indicate that the proposed framework is theoretically highly feasible and demonstrates good applicability and efficiency in practical production. The constructed blockchain network meets the actual needs of responsibility management in smart city BIM projects, enhancing the transparency and reliability of project management.

As critical nodes in rural and urban development, small towns play a significant role in promoting regional economic growth. However, challenges such as unreasonable spatial layouts and unclear industrial positioning have emerged, hindering sustainable development for both the towns and their surrounding areas. To elucidate the relationship between tropical island towns and regional development, this study examines 100 representative towns in Hainan Province, China. Utilizing ArcGIS software and spatial analysis methods—including kernel density estimation, nearest neighbor index, and spatial autocorrelation analysis—the spatial distribution patterns of these towns and their correlations with regional development factors are explored. The findings reveal significant clustering in the spatial distribution of towns, particularly tourist towns, which tend to concentrate in areas rich in tourism resources. Key factors influencing town development are identified, such as resource endowment, market accessibility, transportation infrastructure, and population distribution. These insights provide theoretical guidance for optimizing the spatial layout of small towns by leveraging geographical and resource advantages, thereby enhancing sustainable development and regional competitiveness.

This study investigates the effects of limited Chinese pinyin learning on American elementary school students' language performance, with particular emphasis on potential cross-linguistic interference and the benefits of pinyin learning. Participants from two schools, one receiving Chinese instruction including pinyin (experimental group) and one with no Chinese instruction (control group), in first and fourth grades were assessed on measures of Nonverbal Ability (NA), Phonological Awareness (PA), the Peabody Picture Vocabulary Test (PPVT-4), Chinese Fluency Test (CFT), Commonly Mispronounced Chinese Words Read Aloud Test (CMCT), and Commonly Mispronounced English Words Read Aloud Test (CMET). Results indicated that the modest exposure to pinyin, delivered as five 30-min sessions per three weeks, did not significantly interfere with the acquisition of English vocabulary, as evidenced by comparable performance on PPVT-4 and CMET across groups. In first grade, CMCT was positively correlated with NA, PA, and CMET, while in fourth grade, significant correlations were observed only with CMET and CFT. Furthermore, error analyses revealed that the majority of mis- pronunciations were attributable to intrinsic properties of the English language rather than to negative transfer from pinyin instruction. These differential patterns suggest that a developmental shift in how Chinese pinyin interacts with students' cognitive and linguistic abilities across grade levels, the most effective period for introducing pinyin instruction appears to be around the first grade, when phonological systems are still developing, while by fourth grade, when these systems are largely consolidated, the cognitive benefits of pinyin learning become less direct, although early pinyin skills then appear to support vocabulary acquisition.