Fan Zhou

3D Gaussian Splatting (3DGS) enables real-time photorealistic scene reconstruction, yet its segmentation tasks suffer from two critical flaws: poor 3D consistency (e.g., blurred instance boundaries and unstable cross-view semantic association) and insufficient structural awareness near ambiguous object boundaries. To address these issues, this paper proposes a semantic-guided 3D Gaussian segmentation framework. We first process multi-view RGB images and semantic masks to generate 2D semantic codes and initial Gaussian parameters. We then build a Gaussian-level joint representation by combining flow-aligned appearance cues, CLIP-derived semantic descriptors, and explicit Gaussian geometry. Semantic-guided adaptive decomposition identifies boundary-sensitive Gaussians through cross-view boundary statistics and semantic uncertainty, while multi-view voting further refines instance boundaries. Finally, 3D global optimization unifies instance association and labeling. Experiments on ScanNet and SPIn-NeRF show improved results under the reported evaluation protocol and the stated subset definitions. In particular, on the ScanNet benchmark, compared with the evaluated 3DGS baseline InstanceGaussian, SG-AGS improves mAcc@0.25 by 16.2 points for category-agnostic 3D instance segmentation and by 25.6 points for text-query-based open-vocabulary labeling of segmented 3D instances. These results support the usefulness of SG-AGS on the two evaluated benchmarks and suggest potential value for object-level scene querying and digital-twin inspection, while broader generalization to external captures and other scene domains still requires further validation.

Three-view 3D Gaussian Splatting is prone to insufficient geometric constraints, overfitting to sparse color observations, and uneven allocation of Gaussian representation capacity, leading to structural distortions and local artifacts. We propose a DNGaussian-based framework that combines branch-decoupled regularization with edge-aware selective densification. Monocular relative depth maps are generated offline using Depth Anything V2 with a ViT-S backbone and incorporated as relative geometric supervision. DropGaussian is applied only to the color-rendering branch, whereas Hard-depth and Soft-depth updates use the complete Gaussian set, preventing stochastic suppression from directly perturbing depth supervision. Sobel responses from the three training views are accumulated as cross-view Gaussian edge statistics and used solely to reduce the effective screen-space gradient threshold for Gaussians with high cumulative mean edge scores; cloning and splitting follow the base DNGaussian operations. Relative to DNGaussian, the proposed method improves mean PSNR by 1.50 dB on LLFF and 1.01 dB on DTU, while maintaining a comparable mean Gaussian count on LLFF. These results demonstrate its effectiveness for three-view novel view synthesis.

Extremely sparse-view 3D Gaussian Splatting (3DGS) often suffers from depth drift, incorrect occlusions, and floating artifacts because supervision is unavailable between training cameras. We propose a stability-guided relative geometry distillation (SG-RGD) framework that extends geometric supervision to interpolated pseudo-views while reducing the influence of unreliable regions. Perturbation-based Pseudo-view Stability (PVS) estimates pixel-wise continuous stability weights from appearance differences between normal and mildly scale-perturbed renderings at the same pseudo-camera. Relative geometry distillation (RGD) uses these weights to modulate scale- and shift-invariant correlation alignment between Gaussian-rendered depth and relative depth predicted by a frozen monocular teacher. Progressive Multi-scale Pseudo-depth Curriculum (PMPC) introduces this supervision in a coarse-to-fine manner. On the nine-scene Mip-NeRF 360 three-view benchmark, SG-RGD improves PSNR from 10.6955 to 10.9600 dB and SSIM from 0.1422 to 0.2150 while reducing LPIPS from 0.7146 to 0.7100 and AVGE from 0.3849 to 0.3716 compared with DNGaussian. Experiments on Mip-NeRF 360 six-view and NeRF Synthetic eight-view further characterize performance across view coverage and data distributions, while ground-truth geometry diagnostics show that lower stability weights tend to correspond to larger geometric errors. All proposed modules are training only, preserving the standard 3DGS inference pipeline.

To address the limitations of traditional single-point detection methods in monitoring the sulfur content of ship fuel (FSC), which are inadequate in meeting the regulatory demands of high-traffic ports, this study proposes an integrated analytical approach based on artificial intelligence. This approach synthesizes multi-source heterogeneous data, including historical fuel testing records, Automatic Identification System (AIS) trajectory data, ship and operator profiles, technical specifications, fuel supply chain documentation, fundamental ship attributes and so on. Following rigorous data cleaning and preprocessing procedures, a refined dataset comprising 3046 records collected between 2017 and 2024 from the Port of Ningbo was utilized. Initially, multiple linear regression analysis was con-ducted to identify key factors influencing sulfur emissions, resulting in an R² value of 0.67. Based on these findings, a deep neural network model was developed using TensorFlow to enable real-time estimation of FSC and classification of compliance risk levels. The results indicate that the proposed method exhibits high estimated accuracy and robustness. An AI-based intelligent monitoring module, developed based on this research, has been integrated into the ship exhaust gas detection system at the Port of Ningbo. This module enables real-time analysis of inbound ships and intelligent identification of potentially non-compliant ships, thereby significantly improving the precision and efficiency of port regulatory operations. This study not only contributes to the theoretical framework for ship fuel compliance monitoring but also provides a practical and scalable technical solution for intelligent port governance.

Reservoirs face growing challenges in safety and sustainable management, requiring systematic approaches that integrate monitoring, analysis, and decision support. To address this need, this study develops an integrated information system framework with a four-layer architecture, encompassing “perception,” “data,” “model,” and “application.” The perception layer establishes a multi-platform monitoring network based on fused multi-sensor data. The data layer manages heterogeneous information through correlation mechanisms at the physics, semantics, and application levels. The model layer supports decision-making through a cross-coupled analytical framework for the coordinated management of water safety, resources, environment, and ecology. Finally, the application layer utilizes virtual-physical mapping and dynamic reasoning to implement a closed-loop management system encompassing forecasting, warning, simulation, and planning. This framework was implemented and validated at the Ye Fan Reservoir in Hubei Province, China. By integrating components like “One Map,” flood dispatching, safety monitoring, early warning, video surveillance, and operational supervision, a three-dimensional perception network was constructed. This deployment significantly improved the precision, reliability, and scientific basis of reservoir operation. The integrated monitoring and management system presented in this paper, driven by heterogeneous sensor networks, provides an effective and generalizable solution for modern reservoir management, with the potential for extension to broader water resource and infrastructure systems.

Ship emissions represent a significant challenge to environmental sustainability and public health. The widely implemented Emission Control Area (ECA) policy aims to mitigate these emissions; however, existing research often overlooks inland and underdeveloped coastal regions. This study evaluates the impact of China’s Domestic Emissions Control Area (DECA) policy on SO₂ and PM_2.5 air concentrations in three inland cities along the Yangtze River and three underdeveloped coastal cities since its inception in 2018. Employing regression discontinuity (RD) and difference-in-differences (DID) methodologies, this analysis seeks to provide a comprehensive understanding of the DECA’s effects on air quality and its implications for sustainability in these areas. This study revealed that DECA policies resulted in significant improvements in both SO₂ and PM_2.5 concentrations, contributing directly to environmental sustainability and potential public health co-benefits in inland urban areas as well as underdeveloped coastal cities. In the case of inland cities, the daily average concentrations of SO₂ in Yichang and Taicang decreased by 5.3% and 18.9%, respectively, while the average daily concentrations of PM_2.5 saw reductions of 21.9% and 13.9%, respectively. Among the underdeveloped coastal cities, the average daily concentration of SO₂ in Haikou, Danzhou, and Sanya declined by 1.6%, 20.2%, and 21.2%, respectively; additionally, the average daily concentrations of PM_2.5 in Danzhou and Sanya decreased by 13.8% and 9.5%, respectively. The effectiveness of this policy exhibited geographical variation within inland cities and was influenced by urban development indicators in coastal areas. These findings not only underscore the success of the DECA policy in enhancing air quality but also highlight its role in advancing sustainable development goals. They provide essential evidence for formulating effective and sustainable emissions control strategies applicable to similar inland regions and underdeveloped coastal settings worldwide.

Limiting the fuel sulphur content (FSC) of ships can significantly reduce the harm caused by ship emissions, and analyzing ship exhaust gas to estimate FSC is a rapid, efficient, and low-cost monitoring method. To solve the difficulty in measuring FSC using sniffer equipment in a complex port area, a ship emission monitoring option for FSC measurement in complex environments is proposed here. First, the exhaust gas measurement data of a time series collected using the sniffer equipment were examined to determine the dataset that could be used to estimate FSC. Second, the background value of polluted gases in the environment was dynamically calculated to suppress the interference of various pollution sources. The gas-measured value series was then converted into a mean value series, and the peak points in the mean value series were automatically selected for the calculation of FSC. Finally, the wind speed, wind direction, automatic identification system information, plume diffusion model, and FSC results of ship targets around the equipment were correlated. Between June and August 2023, we conducted a field observation campaign at Ningbo Port, China, where 2624 ships were monitored. A comparison of the real and measured FSC values of eight ships showed that the system could accurately measure FSC at 0.10% (m/m) and 0.50% (m/m) levels despite measurement uncertainty that may be greater at a 0.01% (m/m) FSC level. The FSC statistics of 2624 ships showed that the FSC of small seagoing ships was relatively higher than that of other types of ships. This study proposes a monitoring option for ship emissions, designs and develops an associated system, and collects data to validate the effectiveness and accuracy of this option. This approach has been integrated into daily business operations within the maritime sector, significantly enhancing the efficiency of supervision in this field.

Inland river ships (IRSs) use diesel with a lower sulfur content and emit relatively low emissions, making it challenging to monitor their emissions. Sniffer monitoring equipment was installed from August 2020 to June 2022 at the Gezhou Dam of the Yangtze River and monitored emissions from 8,238 IRSs passing through the lock. We partnered with the maritime department to select 100 ships passing through the lock to extract fuel oil samples for direct fuel sulfur content (FSC) detection, which determined the actual FSC of the passing ships. The monitoring data from these 100 ships indicated that the relative error of the SO₂ emission factors (EFs) and FSC results is significant at the 10-parts-per-million level. The monitoring data from the remaining 8,138 ships showed that the EFs of NO, NO₂, PM_2.5, and PM₁₀ were 24.02 ± 16.92 g kg⁻¹, 10.30 ± 18.08 g kg⁻¹, 0.72 ± 0.60 g kg⁻¹, and 0.92 ± 0.70 g kg⁻¹, respectively. The NOx EFs of container ships are higher than those of other ship types, while the PM EFs for different ship types do not significantly differ. Based on these EFs, we calculated the average emission rates for different types of ships passing through locks, which is a real-time measurement method for estimating ship emissions. In addition, a comparison of ship EF measurements over the past 20 years revealed that EF values for SO₂, NOx, and PM exhibited a downward trend, with the calculated results of the current study determined to be the lowest numerical level.

The implementation effectiveness of ship emission control area (ECA) policies can be effectively evaluated using econometric models. However, existing studies mainly focus on changes in SO₂ concentrations in the air. In order to comprehensively assess the impact of ECA policies on air quality, this study takes Ningbo Port in China as an example and uses a regression discontinuity (RD) model to analyze the influence of ship emissions around the wharf on concentrations of SO₂, NO₂, and particulate matter (PM) in the air. The results indicate that individual ships’ activities within the monitoring area (within 300 m) make a relatively small contribution to the concentration of SO₂ in the air and do not form a significant breakpoint. However, there is a noticeable breakpoint in the concentration of NO₂ around the monitoring point as the ship approaches. At the same time, the variation range of PM_2.5 is significantly greater than that of PM₁₀, which aligns with the characteristics of PM emitted by ships. The experimental results have passed three robustness tests, demonstrating that the current policy on ship ECAs has a positive limiting effect on SO₂ emissions and, to some extent, reduces PM emissions. However, further reductions in ship emissions may require more restrictions in nitrogen oxide emissions.