Search Results (67)

Remote sensing objects often exhibit significant scale variations, high aspect ratios, and diverse orientations. The anisotropic spatial distribution of such objects’ features leads to the conflict between feature representation and boundary regression caused by the coupling of different attribute parameters: previous detection methods based on square-kernel convolution lack the overall perception of large-scale or slender objects due to the limited receptive field; if the receptive field is simply expanded, although more context information can be captured to help object perception, a large amount of background noise will be introduced, resulting in inaccurate feature extraction of remote sensing objects. Additionally, the extracted features face issues of feature conflict and discontinuous loss during parameter regression. Existing methods often neglect the holistic optimization of these aspects. To address these challenges, this paper proposes SODE-Net as a systematic solution. Specifically, we first design a multi-scale fusion and spatially orthogonal convolution (MSSO) module in the backbone network. Its multiple shapes of receptive fields can naturally capture the long-range dependence of the object without introducing too much background noise, thereby extracting more accurate target features. Secondly, we design a multi-level decoupled detection head, which decouples target classification, bounding-box position regression and bounding-box angle regression into three subtasks, effectively avoiding the coupling problem in parameter regression. At the same time, the phase-continuous encoding module is used in the angle regression branch, which converts the periodic angle value into a continuous cosine value, thus ensuring the stability of the loss value. Extensive experiments demonstrate that, compared to existing detection networks, our method achieves superior performance on four widely used remote sensing object datasets: DOTAv1.0, HRSC2016, UCAS-AOD, and DIOR-R. Full article

The adoption of cryptocurrency as a payment instrument by firms has sparked ongoing debates about how such strategic moves are perceived by key stakeholders. This study investigates how investors react when an e-commerce firm adds or withdraws from providing cryptocurrency as a payment option. To explore these aspects, we examine two cases: MercadoLibre’s decision to introduce Meli Dólar as a payment option, representing the inclusion of cryptocurrency, and eBay’s withdrawal from the Libra project, representing strategic exclusion. We assess the causal impact of these strategies by employing a Regression Discontinuity Design (RDD) and deriving the observation period by using an optimal bandwidth method. The results indicate that there was an immediate decline in share prices following the adoption of the Meli Dólar as a payment instrument and an immediate increase following the decision to withdraw from using Libra as a payment instrument. The findings suggest that including cryptocurrency as a payment method may run counter to investor expectations. This study contributes to the discourse on the viability of cryptocurrency adoption by e-commerce firms and emphasizes the importance of understanding how decisions around cryptocurrency convey market signals, which may have strategic implications for a firm’s overall strategy. Full article

Oriented ship detection in Synthetic Aperture Radar (SAR) images has broad applications in maritime surveillance and other fields. While deep learning advancements have significantly improved ship detection performance, persistent challenges remain for existing methods. These include the inherent misalignment between regression and classification tasks and the boundary discontinuity problem in oriented object detection. These issues hinder efficient and accurate ship detection in complex scenarios. To address these challenges, we propose TIAR-SAR, a novel oriented SAR ship detector featuring a task interaction head and composite angle regression. First, we propose a task interaction detection head (Tihead) capable of predicting both oriented bounding boxes (OBBs) and horizontal bounding boxes (HBBs) simultaneously. Within the Tihead, a “decompose-then-interact” structure is designed. This structure not only mitigates feature misalignment but also promotes feature interaction between regression and classification tasks, thereby enhancing prediction consistency. Second, we propose a joint angle refinement mechanism (JARM). The JARM addresses the non-differentiability problem of the traditional rotated Intersection over Union (IoU) loss through the design of a composite angle regression loss (CARL) function, which strategically combines direct and indirect angle regression methods. A boundary angle correction mechanism (BACM) is then designed to enhance angle estimation accuracy. During inference, BACM dynamically replaces an object’s OBB prediction with its corresponding HBB if the OBB exhibits excessive angle deviation when the angle of the object is near the predefined boundary. Finally, the performance and applicability of the proposed methods are evaluated through extensive experiments on multiple public datasets, including SRSDD, HRSID, and DOTAv1. Experimental results derived from the use of the SRSDD dataset demonstrate that the mAP50 of the proposed method reaches 63.91%, an improvement of 4.17% compared with baseline methods. The detector achieves 17.42 FPS on 1024 × 1024 images using an RTX 2080 Ti GPU, with a model size of only 21.92 MB. Comparative experiments with other state-of-the-art methods on the HRSID dataset demonstrate the proposed method’s superior detection performance in complex nearshore scenarios. Furthermore, when further tested on the DOTAv1 dataset, the mAP50 can reach 79.1%. Full article

Objectives: The objective of this single-site retrospective study was to assess the association between Computed Tomography (CT)-based whole-body composition values with dose reduction in patients with a diagnosis of epithelial ovarian cancer (EOC) treated with poly ADP-ribose polymerase (PARP) inhibitors (PARPi). Methods: Forty-eight patients (median age 61 years; interquartile range 53–68.5) with EOC who had a thorax and abdomen CT scan (performed before starting PARPi) were enrolled. Recorded clinical data included age, weight, height, stage, start and end date of PARPi, dose reduction, premature discontinuation of therapy, date of last contact, progression, and death. Body composition values were automatically extracted by dedicated software. Given the exploratory nature of the study, the statistical analysis combined univariate assessments (univariate logistic regression) used to evaluate the individual effect of each variable on the probability of dose reduction, with a classification tree approach—a data-driven machine learning method considering all variables simultaneously as covariates. This integrated strategy was designed to identify empirical cut-offs defining body composition profiles associated with increased risk of toxicity. Results: Univariate logistic regression showed no statistically significant effect of body composition variables on the probability of dose reduction. Due to the complexity of variable relations, a machine-learning approach with a classification tree showed that SKM (skeletal muscle) was the sole body composition variable significantly associated with dose reduction. Specifically, there was a higher risk of dose reduction with SKM values ≥ 7506 cm³ and < 8650 cm³ (p = 0.0118). Conclusions: In this exploratory study, a significant association of whole-body composition parameters (SKM) with dose reduction was observed in patients with a 7506 cm³ ≤ SKM < 8650 cm³. If confirmed in larger cohorts, these findings could help clinicians identify patients who might benefit from an upfront reduced PARPi dose. Full article

The distribution of in situ stress fields in reservoirs is critical for the accurate exploration and efficient exploitation of hydrocarbon resources, especially in deep, fault-developed strata where tectonic activities significantly complicate stress field patterns. To clarify the perturbation effects of faults on in situ stress fields in deep reservoirs, this study combines dynamic–static parameter conversion models derived from laboratory experiments (acoustic emission Kaiser effect and triaxial compression tests) with a coupled “continuous matrix–discontinuous fault” numerical framework implemented in FLAC^3D6.0. Focusing on the BKQ Formation reservoir in the MH area, China, we developed a multivariate regression-based inversion model integrating gravitational and bidirectional tectonic stress fields, validated against field measurements with errors of −2.96% to 9.07%. The key findings of this study include the following: (1) fault slip induces stress reductions up to 22.3 MPa near fault zones, with perturbation ranges quantified via exponential decay functions (184.91–317.74 m); (2) the “continuous matrix–discontinuous fault” coupling method resolves limitations of traditional continuum models by simulating fault slip through interface contact elements; and (3) stress redistribution exhibits NW-SE gradients, aligning with regional tectonic compression. These results provide quantitative guidelines for optimizing hydrocarbon development boundaries and hydraulic fracturing designs in faulted reservoirs. Full article

Synthetic Aperture Radar (SAR) is a remote sensing technology that can realize all-weather and all-day monitoring, and it is widely used in ocean ship monitoring tasks. Recently, many oriented detectors were used for ship detection in SAR images. However, these methods often found it difficult to balance the detection accuracy and speed, and the noise around the target in the inshore scene of SAR images led to a poor detection network performance. In addition, the rotation representation still has the problem of boundary discontinuity. To address these issues, we propose S4Det, a Sinusoidal Single-Stage SAR image detection method that enables real-time oriented ship target detection. Two key mechanisms were designed to address inshore scene processing and angle regression challenges. Specifically, a Breadth Search Compensation Module (BSCM) resolved the limited detection capability issue observed within inshore scenarios. Neural Discrete Codebook Learning was strategically integrated with Multi-scale Large Kernel Attention, capturing context information around the target and mitigating the information loss inherent in dilated convolutions. To tackle boundary discontinuity arising from the periodic nature of the target regression angle, we developed a Sine Fourier Transform Coding (SFTC) technique. The angle is represented using diverse sine components, and the discrete Fourier transform is applied to convert these periodic components to the frequency domain for processing. Finally, the experimental results of our S4Det on the RSSDD dataset achieved 92.2% mAP and 31+ FPS on an RTXA5000 GPU, which outperformed the prevalent mainstream of the oriented detection network. The robustness of the proposed S4Det was also verified on another public RSDD dataset. Full article

Optical sensor systems are essential for space target detection. However, previous studies have prioritized detection accuracy over model efficiency, limiting their deployment on resource-constrained sensors. To address this issue, we propose the lightweight space target real-time detection transformer (STar-DETR), which achieves a balance between model efficiency and detection accuracy. First, the improved MobileNetv4 (IMNv4) backbone network is developed to significantly reduce the model’s parameters and computational complexity. Second, group shuffle convolution (GSConv) is incorporated into the efficient hybrid encoder, which reduces convolution parameters while facilitating information exchange between channels. Subsequently, the dynamic depthwise shuffle transformer (DDST) feature fusion module is introduced to emphasize the trajectory formed by space target exposure. Finally, the minimum points distance scylla intersection over union (MPDSIoU) loss function is developed to enhance regression accuracy and expedite model convergence. A space target dataset is constructed, integrating offline and online data augmentation techniques to improve robustness under diverse sensing conditions. The proposed STar-DETR model achieves an $A{P}_{0.5:0.95}$ of 89.9%, successfully detecting dim and discontinuous streak space targets. Its parameter count and computational complexity are reduced by 64.8% and 41.8%, respectively, highlighting its lightweight design and providing a valuable reference for space target detection in resource-constrained optical sensors. Full article

Automotive exhaust emissions contribute significantly to air pollution in developing countries. However, the effectiveness of driving restriction policies (DRPs) is unclear, and most research on China emphasizes Beijing. This study used Chengdu, a typical large city in China, to examine the impact of a DRP on air quality. To alleviate potential endogeneity threats, we employed a regression discontinuity design to verify the policy’s effect. The results show that the DRP significantly reduced air pollution levels, effectively improving air quality in restricted areas. The heterogeneity analysis found that (1) the DRP effectively reduced pollution in newly added and original areas, while the air quality in adjacent areas deteriorated; and (2) the DRP significantly improved air quality during peak travel periods but had no significant impact in other periods. Our results indicate that the DRP is an effective tool for urban environmental governance but presents potential negative aspects. Therefore, restricted areas and periods should be carefully considered when designing similar policies. This study provides significant insights into the governance of automotive exhaust emissions pollution for large cities in developing countries. Full article

This study investigates the effect of the low emission zone (LEZ), designed to restrict old diesel vehicles, on air quality in Seoul, Republic of Korea, using the regression discontinuity in time (RDiT) approach. While previous studies have examined LEZ impacts using traditional econometric models such as time series and panel data approaches, our research uniquely integrates high-frequency daily weather data to better control for confounding environmental variables and captures time-of-day effects on pollutant concentrations. Our findings reveal that the LEZ policy effectively reduced NO₂ and SO₂ concentrations by 4.7% and 11.6%, respectively. Notably, during daytime hours, when traffic is heaviest, NO₂, SO₂, and PM₁₀ concentrations decreased by 7.1%, 14.8%, and 13.6%, respectively. These results suggest that the observed improvements can be attributed not only to reduced diesel vehicle registrations but also to significant declines in overall traffic volume. Full article

Abstract: Urban greenspaces have been essential in supporting residents’ well-being during the COVID-19 pandemic, particularly under strict lockdown measures. However, the impact of changing containment policies on residents’ perceived greenspace quality remains insufficiently explored. This study utilized online survey data collected between 11 October and 29 December 2022, in Shanghai, coinciding with the major policy shift on 5 December 2022. A probability proportionate to size sampling was adopted to survey residents aged 18 and above who had lived in the city for at least six months, yielding a total of 577 valid responses. We assessed residents’ perceived greenspace quality using 20 park- and community-level variables, focusing on both overall quality and specific features of greenspaces. A regression discontinuity design (RDD) was applied to evaluate how the lifting of the COVID-19 policies influenced residents’ perceptions of parks and community greenspaces. Our RDD estimation indicates no statistically significant change in residents’ overall perceived quality of parks after the policy shift, except for increased satisfaction with specific features such as plant diversity, maintenance, seating areas, trails, and large open spaces. In contrast, residents who responded after the policy shift reported a significantly higher perceived quality of community greenspaces compared to those who completed the survey before the shift (0.609 score difference, p < 0.01), with notable increases in satisfaction regarding plant diversity, maintenance, and seating areas. Perception of plant quantity remained unchanged in both types of greenspaces. Residents expressed greater satisfaction with sports facilities in parks, while community greenspaces were preferred for their water features and esthetic qualities. By adopting an RDD with a unique dataset, this study contributes empirical evidence to the current ongoing debate on the role of urban greenspace during the later stages of COVID-19. Specifically, it examines how changes in public health policy and the resulting increase in mobility might affect residents’ perceived greenspace quality. The findings can assist decision-makers and urban planners in developing more adaptive strategies to address the diverse needs of residents for greenspaces during the transitional period of a public health crisis. Full article

The Comprehensive Commercial Logging Ban Policy in all natural forests (CCLBP) as the strictest forest conservation measure brings uncertainty to the income of farmers engaged in forest land management. Therefore, clarifying the impact and heterogeneity of the CCLBP on farmers’ income has become a significant issue of current concern. Based on county-level panel data from China covering the period 2000–2020, this study uses Regression Discontinuity Design (RDD) to identify the impact of the CCLBP on farmers’ income. The empirical results show that (1) the CCLBP has a significantly positive effect on farmers’ income, with the policy leading to an increase in farmers’ income of approximately RMB 411–582; (2) the impact of the CCLBP on farmers’ income exhibits regional heterogeneity, with significant positive effects observed in Hebei, Shandong, Hubei, and Shaanxi, significant negative effects observed in Guangxi, and insignificant effects observed in other provinces; and (3) the CCLBP not only promotes the development of non-agricultural industries and labor mobility but also effectively reduces capital outflow, thereby increasing farmers’ income. This study contributes to the understanding of the underlying mechanisms between the CCLBP and farmers’ income, and it has significant practical implications for promoting the increase in farmers’ income, narrowing the income gap among farmers, and achieving common prosperity. It can also provide valuable insights and guidance for global forest protection. Full article

The COVID-19 pandemic has led to significant transformations in industries globally, particularly those heavily reliant on human interaction, such as the event industry. However, the effects of COVID-19 on the event industry have not been thoroughly explored in previous studies. This study utilizes secondary data from the Korean Statistical Information Service, covering 16 cities and regions from 2018 to 2022, to analyze the effects of COVID-19 on the event industry and how the pandemic has reshaped the sector’s landscape and sustainability. We employed a Regression Discontinuity Design (RDD) model to assess the causal impact and utilized Garthwaite’s (2014) Dynamic Discontinuity model to explore the dynamic effects over time. The results demonstrate that, initially, COVID-19 had a considerable disruptive influence on the event industry, severely affecting face-to-face interactions and operations. However, our findings reveal significant signs of adaptation and recovery in the industry by 2022, with the initial negative impacts no longer evident. This study highlights the event industry’s resilience, the progressive nature of its post-pandemic recovery, and its path toward sustainable practices in a post-pandemic era. Full article

This study introduces a novel satellite image digital surface model (DSM) reconstruction framework grounded in deep learning methodology. The proposed framework effectively utilizes a rational polynomial camera (RPC) model to establish the mapping relationship between image coordinates and geographic coordinates. Given the expansive coverage and abundant ground object data inherent in satellite images, we designed a lightweight deep network model. This model facilitates both coarse and fine estimation of a height map through two distinct stages. Our approach harnesses shallow and deep image information via a feature extraction module, subsequently employing RPC Warping to construct feature volumes for various angles. We employ variance as a similarity metric to achieve image matching and derive the fused cost volume. Following this, we aggregate cost information across different scales and height directions using a regularization module. This process yields the confidence level of the current height plane, which is then regressed to predict the height map. Once the height map from stage 1 is obtained, we gauge the prediction’s uncertainty based on the variance in the probability distribution in the height direction. This allows us to adjust the height estimation range according to this uncertainty, thereby enabling precise height value prediction in stage 2. After conducting geometric consistency detection filtering of fine height maps from diverse viewpoints, we generate 3D point clouds through the inverse projection of RPC models. Finally, we resample these 3D point clouds to produce high-precision DSM products. By analyzing the results of our method’s height map predictions and comparing them with existing deep learning-based reconstruction methods, we assess the DSM reconstruction performance of our proposed framework. The experimental findings underscore the robustness of our method against discontinuous regions, occlusions, uneven illumination areas in satellite imagery, and weak texture regions during height map generation. Furthermore, the reconstructed digital surface model (DSM) surpasses existing solutions in terms of completeness and root mean square error metrics while concurrently reducing the model parameters by 42.93%. This optimization markedly diminishes memory usage, thereby conserving both software and hardware resources as well as system overhead. Such savings pave the way for a more efficient system design and development process. Full article

To protect forest land from loss and mitigate the global climate crisis, China has proposed a stringent natural forest protection plan, known as China’s natural forest logging ban (NFLB). This policy aims to halt the over-exploitation of natural forests, restore forest ecosystem functions, and promote regional green economic development. This study uses a regression discontinuity design (RDD) model to quantitatively and comprehensively assess the effectiveness of this policy in the key state-owned forest regions in Northeast China. Additionally, it analyzes the heterogeneity and structural characteristics of the policy’s effects on the internal composition of ecological and economic systems. The empirical results are as follows: (1) Ecological and economic impacts: The policy has successfully achieved its ecological objectives by significantly enhancing the quality and value of ecosystem services. However, it has also had a notable adverse impact on economic development, particularly in the timber supply sector, reducing the conversion efficiency of ecosystem service values into economic benefits. (2) Structural analysis: The logging ban effectively promoted the value of various ecosystem services, particularly enhancing regulatory and support functions, with a LATE estimate of approximately 8.47 units. The implementation of the policy caused a negative growth in the output value of supply-oriented ecological products, and the significance level was lower than 0.1. Conversely, the LATE estimates for different types of GDP indicate a negative growth in supply-type GDP due to the policy, with p < 0.1. (3) Heterogeneity: On the one hand, a simplistic and singular approach to logging prohibition may constrain the efficiency of enhancing ecosystem service values. On the other hand, although the policy disrupted the majority of traditional forest enterprise operations, business models focusing on quality and technology improvements were able to mitigate this impact. Full article

Regional integration plays a pivotal role in the socio-economic advancement of various global regions and is closely linked with the expansion of construction land. This expansion is a major contributor to urban carbon emissions. Utilizing a geographical regression discontinuity design (GRDD), this paper estimates the impact of urban construction land expansion on carbon emissions and explores the underlying mechanisms within the regional integration process of the Yangtze River Delta (YRD), China. The findings reveal that urban construction land expansion significantly influences carbon emissions, displaying an inverted “U”-shaped pattern. Furthermore, this expansion affects carbon emissions through the transformation of industrial structures, shifts in consumption patterns, and enhancements in scientific and technological investments. Our findings span the performance of the Yangtze River Delta from its early development stages to a relatively mature phase. This paper also partially reveals how the Yangtze River Delta, with both megacities and large- to medium-sized cities, manages urban construction land expansion during the integration process and strives for low-carbon emissions reduction. These results can provide green growth recommendations that balance socio-economic development, low-carbon emissions, and social equity not only for other urban agglomerations in China but also for similar regions in other developing countries by altering construction land utilization patterns. Full article