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.

This paper aims to explore the effect of teacher–student collaboration on academic innovation in universities in different stages of collaboration.

Based on collaboration life cycle, this paper divided teacher–student collaboration into initial, growth and mature stages to explore how teacher–student collaboration affects academic innovation.

Collecting data from National Science Foundation of China, the empirical analysis found that collaboration increases the publication of local (Chinese) papers at all stages. However, teacher–student collaboration did not significantly improve the publication of international (English) papers in the initial stage. In the growth stage, teacher–student collaboration has a U-shaped effect on publishing English papers, while its relationship is positive in the mature stage.

The results offer suggestions for teachers and students to choose suitable partners and also provide some implications for improving academic innovation.

This paper constructed a model in which the effect of teacher–student collaboration on academic innovation in universities was established.

Axleboxes are an important structure that connects the wheelset with the bogie frame. Inside axlebox bogies have lower wheelset yaw angle stiffness and better curve-passing performance. The purpose of this paper to study the differences in the wear evolution law and the influencing factors of the two types of metro vehicles.

This paper established the dynamic model and wear model of both outside axlebox and inside axlebox metro vehicles to research the wheel wear evolution law of the two types of vehicles. The curve passing performance of two vehicles is analyzed. The effect of key parameters on wheel wear is studied, including the lateral distance of the axlebox, the longitudinal stiffness of the rotary arm node, the lateral stiffness of the rotary arm node and the wheel profiles.

The results showed that the model of inside axlebox metro vehicles improved vehicle safety and curve-passing performance. At the same time, inside axlebox metro vehicles reduce wheel wear of the wheel tread area and wheel flange area. When the S1002 wheel tread profile matched with the vehicle parameters, the wheel wear is minimized.

This paper established a dynamic model for inside axleboxes metro vehicles, then used a wheel wear model to analyze the evolution of wheel wear and the key influencing factors of the inside axleboxes metro vehicles.

The peer review history for this article is available at: https://publons.com/publon/10.1108/ILT-07-2024-0256/

Unpredictable economic landscapes have led to a continuous escalation in global economic policy uncertainty (EPU). Improving risk management and sustainability in an environment with high macro risk is critical for business development. This study aims to explore the impact of corporate sustainable development on corporate tax risk.

After using a sample of companies that were A-share listed on the Shanghai and Shenzhen stock exchanges from 2011 to 2021, this paper applies ordinary least squares and a moderate effect model.

Better environmental, social and governance (ESG) performance can weaken corporate tax risk by improving green innovation capability, reputation and information transparency. Meanwhile, the restraining effect of ESG on tax risk was more significant amid high EPU. These impacts were amplified amid higher market competition, lower tax supervision and a lower degree of corporate digital transformation.

The findings emphasize the need for the government to establish a healthy business and tax environment so that enterprises can improve sustainable development and increase their risk management abilities, especially post-COVID-19.

This study guides enterprises and the entirety of society to in paying attention to and promoting ESG practices, which can enhance enterprise tax management.

This study expands the research on the economic consequences of sustainable development and the factors influencing corporate tax risk and EPU.

This study aims to examine the determinants of disclosing repetitive year-to-year key audit matters (KAMs) content by auditors for a client and whether repetitive KAMs content is indicative of lower audit effort.

The authors use a sample of publicly listed firms from New Zealand between 2016 and 2020. Multivariate regression models are used to test the determinants of the extent of repetitive content in the KAMs section of the audit report. The authors compare the KAMs disclosed in the current period to those disclosed in prior period(s) to determine the level of recurring items and repetitive textual content.

The authors find evidence of repetitive KAMs content being disclosed at the client level since the reporting requirement was introduced. In multivariate analyses, the authors find that client business risk and auditor reputation are negatively associated with auditors’ reporting of repetitive KAMs. Furthermore, the authors find that auditors exert lower effort on audits for which they report a higher level of repetitive content in KAMs.

The study provides novel findings that contribute to the literature on auditors’ voluntary reporting of KAMs and provide important practical implications for investors and regulators.

With emerging technologies rapidly changing work processes, it is important to understand the skills and characteristics project managers (PMs) need to effectively manage projects in the digital era. This study determines the underlying competencies needed for digitalization among PMs in the construction industry. The study also identified the most significant competencies needed by PMs in the era of digitalization.

The methodology adopted for the research study was quantitative. It was founded on a thorough review of pertinent literature, which went through a pilot survey study from six project management experts in the construction industry. Based on the comment and feedback, a questionnaire survey was developed and distributed to participants through a convenience sampling technique. The data retrieved were from 100 professional PMs out of 130 questionnaires distributed in the Ghanaian construction industry. Data collected were analyzed using fuzzy synthetic evaluation (FSE).

Based on FSE, the three competency parameters (knowledge, skills, personal characteristics) generated significant indices indicating that all three competencies are significant among construction PMs in the digitalization era. Under the knowledge competency parameter, six sub-competencies comprising a total of 12 variables were identified. For skills, seven sub-competencies consisting of 23 variables were identified. Regarding personal attributes, six sub-competencies with 17 variables were highlighted. Under knowledge, technical knowledge was found to be the most important with an index of 4.212. For skills, leadership skills were rated highest with an index of 4.240. Regarding personal attributes, social or interpersonal skills were deemed most critical with an index of 4.199.

The results provide guidance to both industry and academic stakeholders. For PMs and their employers, the study highlights priority areas for competency development and training related to the era of digitalization. It also informs educational institutions on how to structure project management curricula to best prepare students for jobs of the future. This study gives more insight into the competencies that need more attention for PMs in the digitalization era. As a result, firms that adopt the identified competency will benefit from implementing digitalization in project delivery.

This study makes an original contribution as one of the first to empirically investigate the competencies required of construction PMs in the construction industry in the digitalization era. By focusing on the developing country context of Ghana, the study extends knowledge to an under-researched region and market. It provides a foundation for future comparative research across diverse global contexts.

This study aims to comprehensively analyze the intersection of technology management and innovation management amidst the fourth industrial revolution, uncovering evolving trends and influential contributors.

Using the Bibliometrix R-package, this pioneering research conducts a bibliometric analysis to delve into innovation and technology management literature, quantifying scholarly output and identifying thematic breakthroughs.

The study reveals quantitative insights into the progression of innovation and technology management research, offering guidance on evolving trends, thematic breakthroughs and influential contributors.

The findings offer valuable insights for practitioners and managers, guiding them through emerging trends and recommending a dual focus on fundamental principles and emerging areas for strategic decision-making.

By fostering active engagement with evolving trends, this research contributes to the ongoing technology and innovation management discourse, potentially leading to societal benefits and advancements.

This study pioneers an in-depth bibliometric analysis at the intersection of innovation and technology management, offering unique insights and quantitative assessments of scholarly output and thematic trends, thus adding significant value to the existing literature.