Search results

Vision transformers (ViT) detectors excel in processing natural images. However, when processing remote sensing images (RSIs), ViT methods generally exhibit inferior accuracy compared to approaches based on convolutional neural networks (CNNs). Recently, researchers have proposed various structural optimization strategies to enhance the performance of ViT detectors, but the progress has been insignificant. We contend that the frequent scarcity of RSI samples is the primary cause of this problem, and model modifications alone cannot solve it.

To address this, we introduce a faster RCNN-based approach, termed QAGA-Net, which significantly enhances the performance of ViT detectors in RSI recognition. Initially, we propose a novel quantitative augmentation learning (QAL) strategy to address the sparse data distribution in RSIs. This strategy is integrated as the QAL module, a plug-and-play component active exclusively during the model’s training phase. Subsequently, we enhanced the feature pyramid network (FPN) by introducing two efficient modules: a global attention (GA) module to model long-range feature dependencies and enhance multi-scale information fusion, and an efficient pooling (EP) module to optimize the model’s capability to understand both high and low frequency information. Importantly, QAGA-Net has a compact model size and achieves a balance between computational efficiency and accuracy.

We verified the performance of QAGA-Net by using two different efficient ViT models as the detector’s backbone. Extensive experiments on the NWPU-10 and DIOR20 datasets demonstrate that QAGA-Net achieves superior accuracy compared to 23 other ViT or CNN methods in the literature. Specifically, QAGA-Net shows an increase in mAP by 2.1% or 2.6% on the challenging DIOR20 dataset when compared to the top-ranked CNN or ViT detectors, respectively.

This paper highlights the impact of sparse data distribution on ViT detection performance. To address this, we introduce a fundamentally data-driven approach: the QAL module. Additionally, we introduced two efficient modules to enhance the performance of FPN. More importantly, our strategy has the potential to collaborate with other ViT detectors, as the proposed method does not require any structural modifications to the ViT backbone.

To investigate whether the actual effects of eight drape characteristics of virtual fabrics can be manifested in the Style 3D software.

Image analysis was conducted using MATLAB software to obtain the drape characteristics of virtual fabrics. Pair the drape characteristics of the real and virtual fabrics for difference. The S-W method was used to conduct a normality test to obtain the correlation of paired samples. A paired sample t-test was performed to obtain the significance values.

The simulation restoration performance of the drape coefficient, number of undulations, maximum undulation angle, minimum undulation angle and undulation angle uniformity was good. However, there are differences in the simulation performance of the other three indicators: maximum undulation amplitude, minimum undulation amplitude and undulation amplitude uniformity compared to the drape characteristics of real fabrics.

Provides reference value for the improvement of Style3D software in virtual fabric simulation and finds the main influential parameters and their impact levels that contribute to the realistic representation of virtual fabrics in software. It provides a theoretical basis for course teaching in digital fashion.

Governments worldwide are placing a greater emphasis on enhancing ecology and the environment as a result of escalating ecological issues. One possible approach is sustainable governance. This chapter explores the interrelated roles of internal control, environmental accounting, and environmental auditing mechanisms in promoting sustainable governance and green transformation. By looking at these three aspects, the chapter illustrates how integrated approaches can promote sustainable practices and guarantee adherence to environmental standards. The objective of this chapter is to present a thorough knowledge of the ways in which these components work together to support sustainability as a whole.

This study proposes to conceptualize and explicate the progress of end-to-end data pipelines to enable seamless data flow, addressing the mounting demand for instantaneous information in the vibrant field of marketing analytics. Ensuring timely access to inclusive analytics, it empowers marketers to make conversant decisions promptly. The study espouses a conceptual approach, starting with the classification and identification of various data sources crucial for thorough marketing analytics. It probes into the complexities of designing adaptable and scalable architectures for extensive data pipelines, ensuring effective data collection, processing, and storage. Emphasizing case analyses of marketing analytics tools, the study delivers practical insights and exhaustive examinations of existing solutions. The findings offer a broad understanding of the key components and strategies necessary for realizing smooth data flow in real-time marketing analytics. This investigation contributes expressively to the field by outlining a robust framework for continuous data flow, identifying the benefits and challenges associated with data pipelines. From a hands-on standpoint, it provides organizations with appreciated insights for building effective end-to-end data pipelines, bestowing strategies for implementing real-time marketing analytics outlines that enhance decision-making processes. The originality and value of this research lie in its universal approach to addressing the challenges of real-time marketing analytics, offering a fresh viewpoint on creating seamless data flow frameworks. The gathered information has the latent to drive innovation in marketing analytics, serving as a crucial resource for companies aiming to gain a competitive edge in today’s data-driven market. This inclusive approach ensures that the study’s contributions are both theoretically significant and practically applicable.

Previous studies have rarely integrated the financing modes of a capital-constrained manufacturer with the choices of online sales strategies. To address this gap, the authors study how a manufacturer selects optimal financing modes under different sales strategies in three dual-channel supply chains.

This paper considers three sales strategies, namely, combining a traditional retailer channel with one of the direct selling, reselling and agency selling channels, and two common financing modes, namely, bank financing and retailer financing. The authors obtain equilibrium outcomes of the manufacturer and traditional retailer and then provide the conditions for them to select optimal financing modes under three sales strategies.

The results indicate that the manufacturer’s financing decisions rely on the initial capital and interest rates, and the manufacturer selects retailer financing only if the initial capital is relatively larger. In terms of financing mode options, the retailer financing mode is more beneficial for the manufacturer under the three sales strategies. From the perspective of sales strategies, the direct selling model is more beneficial. In addition, the higher the consumer acceptance of the online channel, the more profits the manufacturer obtains.

This paper provides suggestions on how the capital-constrained manufacturer chooses financing modes and sales strategies.

This paper integrates the financing mode and different sales strategies to investigate the manufacturer’s optimal operational decisions. These sales strategies allow us to investigate the manufacturer’s optimal financing modes in the presence of both different financing modes and sales strategies.

This study aims to investigate the relationship between climate policy uncertainty (CPU) and corporate environmental, social and governance (ESG) performance. We attempt to uncover the underlying rationale of how CPU influences corporate ESG performance and provides empirical evidence for companies’ strategic enhancement of ESG performance with risk reduction objectives.

We conduct a regression analysis using panel data from 4,490 Chinese listed companies spanning the period from 2011 to 2022. In addition, we use propensity score matching analysis (PSM), two-stage least squares (2SLS), system generalized method of moments (sys-GMM) and difference-in-differences (DID) methods to analyze the enterprise systematic risk.

The empirical findings reveal a positive correlation between CPU and corporate ESG performance, with a stronger effect observed in non-state-owned enterprises, heavy-polluting industries and those facing fierce market competition and strict environmental regulation. Mechanism analysis suggests that as CPU increases, companies with higher systemic risk tend to improve ESG performance more significantly, highlighting risk mitigation as a primary motive. Robustness tests further validate the consistency of our conclusions. Additionally, we find that enhancing ESG performance helps mitigate the risks and improve total factor productivity arising from the increased CPU.

This study examines the impact of CPU on the ESG performance of Chinese listed companies and its underlying logic. The conclusions of this paper provide important policy references for coordinated development and security, as well as for effectively mitigating the adverse impact of CPU. We hope to offer insights for companies to identify potential risk factors, thereby enhancing their level of sustainable development and sense of environmental responsibility.

This study presents a systematic literature review (SLR) of the interdisciplinary literature on drones in last-mile delivery (LMD) to extrapolate pertinent insights from and into the logistics management field.

Rooting their analytical categories in the LMD literature, the authors performed a deductive, theory refinement SLR on 307 interdisciplinary journal articles published during 2015–2022 to integrate this emergent phenomenon into the field.

The authors derived the potentials, challenges and solutions of drone deliveries in relation to 12 LMD criteria dispersed across four stakeholder groups: senders, receivers, regulators and societies. Relationships between these criteria were also identified.

This review contributes to logistics management by offering a current, nuanced and multifaceted discussion of drones' potential to improve the LMD process together with the challenges and solutions involved.

The authors provide logistics managers with a holistic roadmap to help them make informed decisions about adopting drones in their delivery systems. Regulators and society members also gain insights into the prospects, requirements and repercussions of drone deliveries.

This is one of the first SLRs on drone applications in LMD from a logistics management perspective.

This study aims to investigate the determinants influencing green mindfulness among university students in Malaysia within the context of the Malaysia 2030 Agenda, focusing on Sustainable Development Goal 17 attainment.

The research uses a quantitative approach with a sample of 203 young adults. It explores the factors of goal difficulty, knowledge and awareness, spirituality, values and perceived university environmental responsibility (PUER), using a novel theoretical framework termed universal identity theory (IT).

Values, knowledge and awareness and PUER significantly contribute to green mindfulness, whereas spirituality and goal difficulty did not exhibit a substantial relationship to green mindfulness.

Limited representation of diverse age groups and the potential influence of seniority on spirituality. Future research should expand the framework to include green behavior and performance, increase the sample size and consider a broader age demographic.

Universities play a crucial role in promoting green mindfulness through the establishment of rules, regulations, environmental initiatives, incentive systems and the introduction of a green mindfulness course. Clear communication channels and top-down approaches are recommended.

This research contributes to understanding the mechanisms that induce green mindfulness among university students in Malaysia, aligning with national and global sustainability goals.

The universal IT provides a comprehensive understanding of how personal, social and community-based identities collectively influence green mindfulness. This theoretical perspective contributes to the environmental psychology and sustainability studies field, offering a culturally sensitive approach.

The purpose of this paper is to investigate the accurate monitoring and assessment of steel bar corrosion in concrete based on deep learning multi-sensor information fusion method. The paper addresses the issue of traditional corrosion assessment models relying on sufficient data volume and low evaluation accuracy under small sample conditions.

A multi-sensor integrated corrosion monitoring equipment for reinforced concrete is designed to detect corrosion parameters such as corrosion potential, current, impedance, electromagnetic signal and steel bar stress, as well as environmental parameters such as internal temperature, humidity and chloride ion concentration of concrete. To overcome the small amount of monitoring data and improve the accuracy of evaluation, an improved Siamese neural network based on the attention mechanism and multi-loss fusion function is proposed to establish a corrosion evaluation model suitable for small sample data.

The corrosion assessment model has an accuracy of 98.41%, which is 20% more accurate than traditional models.

Timely maintenance of buildings according to corrosion evaluation results can improve maintenance efficiency and reduce maintenance costs, which is of great significance to ensure structural safety.

The corrosion monitoring equipment for reinforced concrete designed in this paper can realize the whole process of monitoring inside the concrete. The proposed corrosion evaluation model for reinforced concrete based on Siamese neural network has high accuracy and can provide a more accurate assessment model for structural health testing.

This study aims to investigate the perspective of corporate philanthropy during the coronavirus disease 2019 (COVID-19) in China for firms with various levels of corporate social responsibility (CSR). Specifically, the study appraises the impact of the COVID-19 pandemic on the stock returns and sustainable development of Chinese-listed companies and determines the likelihood of paying donations vis-à-vis firm reputation.

The study used data from 117 Chinese-listed firms engaged in philanthropy during the COVID-19 pandemic. The authors also utilized the stock returns and cash donation data, and owing to the cross-sectional data and continuous nature of dependent variables, they employed the ordinary least squares regression to test the research hypotheses.

The results show that irresponsible actions have a positive relationship with donations. The study particularly reveals that irresponsible firms have significant negative abnormal returns during the first wave of the COVID-19 pandemic.

To the best of our knowledge, this is the first empirical study to explore the perspective of corporate philanthropy during the COVID-19 pandemic for companies with different CSR levels. This study contributes to the empirical research on CSR and provides insights for managerial-cum-financial decisions to encourage managers of irresponsible firms to pursue philanthropic behaviors after crisis events.