Contextual Features Research Articles

LACTNet: A Lightweight Real-Time Semantic Segmentation Network Based on an Aggregated Convolutional Neural Network and Transformer

Transformers have demonstrated a significant advantage over CNNs in modeling long-range dependencies, leading to increasing attention being paid towards their application in semantic segmentation tasks. In the present work, a novel semantic segmentation model, LACTNet, is introduced, which synergistically combines Transformer and CNN architectures for the real-time processing of local and global contextual features. LACTNet is designed with a lightweight Transformer, which integrates a specially designed gated convolutional feedforward network, to establish feature dependencies across distant regions. A Lightweight Average Feature Bottleneck (LAFB) module is designed to effectively capture spatial detail information within the features, thereby enhancing segmentation accuracy. To address the issue of spatial feature loss in the decoder, a long skip-connection approach is employed through the designed Feature Fusion Enhancement Module (FFEM), which enhances the integrity of spatial features and the feature interaction capability in the decoder. LACTNet is evaluated on two datasets, achieving a segmentation accuracy of 74.8% mIoU and a frame rate of 90 FPS on the Cityscapes dataset, and a segmentation accuracy of 71.8% mIoU with a frame rate of 126 FPS on the CamVid dataset.

Approaching information-seeking habits and their contextual features

Introduction. In information seeking, habits manifest themselves in the ways in which people repeatedly and regularly prefer and access certain sources of information. The present study reviews how researchers in diverse field have approached such habits and characterised their contextual features. Method. Qualitative content analysis of a sample of seventy-three studies on the above topic. The research inductively identified three approaches to habits in information seeking and contextual factors affecting habitual information seeking. Results. Many of the investigations on information-seeking habits approach them in terms of the frequency of information source use. Second, such habits are characterised in terms of information-seeking patterns. Third, the information practice approach directs the main attention to habits as integral constituents of people´s customary ways to seek access to information sources as a part of their daily routines. Information-seeking habits are influenced by contextual factors such as the time and place of information seeking, familiarity with information sources, and easy access to information. Conclusion. Information-seeking habits offer familiar and often efficient ways to identify and access information sources. There is a need to explore further the nature of habitual information seeking by conducting longitudinal studies charting people´s ways and motives to prefer similar sources across situations.

FCAN : Speech emotion recognition network based on focused contrastive learning

Traditional speech emotion recognition (SER) models predominantly utilize cross-entropy loss for supervised optimization. Yet, given the limited size of SER datasets, relying solely on cross-entropy loss does not adequately capture the intricacies within the data. Moreover, conventional contrastive learning techniques, though improving performance, do not completely harness the inherent data structure across samples. To surmount these limitations, we introduce an innovative emotion recognition network founded on focused contrastive learning. Our model employs bidirectional gated recurrent unit (GRU) networks to seize critical contextual dependencies. By integrating an attention mechanism, we adeptly merge the contextual features of dialogues with individual utterance attributes, forming a comprehensive multi-stage feature fusion framework. This strategy extracts more elaborate emotional features from the constrained dataset. Additionally, by amalgamating the focal loss function into our training protocol, we mitigate imbalanced sample issues to a degree. We further refine our contrastive loss function by enacting a weighted strategy for intra-dialogue category assignments, where higher penalty weights are ascribed to harder-to-identify categories, thus incentivizing the model to concentrate on distinguishing subtler category differences. This refinement significantly bolsters the model’s proficiency in learning various emotional feature distributions, thereby enhancing its emotion recognition capabilities. Experimental validation on IEMOCAP and MELD benchmark datasets underpins our model’s superiority.

Understanding users' dynamic perceptions of search gain and cost in sessions: An expectation confirmation model

AbstractUnderstanding the roles of search gain and cost in users' search decision‐making is a key topic in interactive information retrieval (IIR). While previous research has developed user models based on simulated gains and costs, it is unclear how users' actual perceptions of search gains and costs form and change during search interactions. To address this gap, our study adopted expectation‐confirmation theory (ECT) to investigate users' perceptions of gains and costs. We re‐analyzed data from our previous study, examining how contextual and search features affect users' perceptions and how their expectation‐confirmation states impact their following searches. Our findings include: (1) The point where users' actual dwell time meets their constant expectation may serve as a reference point in evaluating perceived gain and cost; (2) these perceptions are associated with in situ experience represented by usefulness labels, browsing behaviors, and queries; (3) users' current confirmation states affect their perceptions of Web page usefulness in the subsequent query. Our findings demonstrate possible effects of expectation‐confirmation, prospect theory, and information foraging theory, highlighting the complex relationships among gain/cost, expectations, and dwell time at the query level, and the reference‐dependent expectation at the session level. These insights enrich user modeling and evaluation in human‐centered IR.

Using cloud computing techniques to map the geographic extent of informal settlements in the greater Cape Town Metropolitan Area

Although remote sensing approaches offer unprecedented opportunities to understand urban land cover dynamics including informal settlements areal extent, challenges such as spectral confusions still persist, particularly when segregating land cover types like informal settlements from planned formal settlements. The improvements in Earth Observation (EO) data analytic tools such as introduction of Google Earth Engine (GEE) cloud computing platform, provide prospects to improve separability of these settlements from other urban land cover classes, via their advanced data processing and filtering algorithm, which allows for the synergic use of multisource and multi-temporal data, thus improving detection and monitoring of these settlements. This study harnessed the advance data analytic powers of GEE cloud computing platform coupled with higher resolution Sentinel-2 data to map the geographical extent of informal settlement in the Cape Town Metropolitan Area. The classification yielded six land cover classes: formal settlements, informal settlements, water, bare or built-up areas, vegetated lands, and croplands. Built-up formal settlement was the most dominant class, accounting for 70% of the total Cape Town surface area, while open water was the least dominant, accounting for 2%. Informal settlements accounted for approximately 7% of all settlements. Although overall accuracy was within acceptable limits (68%), some classes, such as vegetated lands and formal settlements, reported low class accuracies due to spectral similarities with other classes. The findings highlight the importance of the GEE platform, as well as the interaction of contextual and spectral characteristics, as well as various sentinel-2 derived water, built up, and vegetation indices in mapping informal settlements. These findings are critical for the facilitation of improved urban planning, provision of services and assisting in alleviating social as well as environmental issues within the Cape Town Metropolitan area.

The intraindividual dynamics of university students' motivation and emotions: The role of autonomy-supportive learning climates and learning activities.

The motivation and emotions of students are context dependent. There are specific moments when students may find their coursework more or less motivating, resulting in stronger or milder emotional responses. Identifying factors directly controllable by teachers empowers them to effectively address challenging situations characterized by lower motivation and increased negative emotions. We aimed to investigate how learning activities and students' perception of teaching practices fostering autonomy relate to competence and value beliefs, and emotions in the context of course participation within higher education. Seventy-seven Taiwanese university students provided 762 learning reports associated with their course participation experiences. The experience sampling method (ESM) was used. Participants responded to ESM surveys on their phones for 14 days, reporting motivational beliefs, emotions and contextual characteristics of the course if they indicated active participation in a course upon receiving notifications from their phones. A significant portion of the variation is attributed to situational fluctuation, suggesting that academic emotions and competence and value beliefs vary within students across measurements. An increase in students' perception of an autonomy-supportive learning climate correlates with higher competence beliefs, intrinsic value and positive emotions, coupled with reduced perceived costs and negative emotions. In contrast to lectures, engaging in independent hands-on work, participating in group collaborative projects or interactive discussions appear to inspire motivation or evoke stronger emotional responses in students. Teachers' teaching practices and classroom learning activities play a pivotal role in shaping students' situational motivation and emotions.

SIIF: Semantic information interactive fusion network for photovoltaic defect segmentation

Any defects or damage on photovoltaic panels significantly reduce photoelectric conversion efficiency and service life. Therefore, the refined defect segmentation technology is the key to the quality inspection of photovoltaic equipment. However, the existing methods ignore the information interaction between multi-levels, causing semantic features to be easily lost and difficult to fuse. In this paper, we propose a segmentation network based on an interactive fusion of semantic information for solar cell defect identification, which can effectively interact with multi-level rich information and enhance the positioning capabilities of high-level semantics. Specifically, we propose a novel Multi-Level Pyramid Strategy (MLPS) to reduce the information gap between multi-level features. MLPS adopts a pooling distribution mechanism to enable multi-level features to interact effectively in the feature pyramid. In addition, to effectively integrate high-level positioning semantics into the neighbor level, we design a new Neighbor Attention Fusion module (NAF) to propagate high-level semantic information gradually. NAF learns spatial neighbor-level similarity combined with high-level attention to balance location information and contextual features. We perform comprehensive experiments to demonstrate the advancement of our approach. Our method achieves 84.3% MIOU on our solar cell dataset, outperforming existing state-of-the-art (SOTA) methods. To verify the generalization ability of our network, we train and evaluate the PSCDE dataset. The results show that our model achieves an excellent 98.35%MIOU.

Target-based building extraction from high-resolution RGB imagery using GEOBIA framework and tabular deep learning model

Extracting building information from remotely sensed images is essential in various geographic and environmental applications. Specifically, there is a growing demand for distinguishing buildings based on roof colors from low-cost but high-resolution RGB images. However, this task presents significant challenges due to the variability in building color, size, and style, as well as their resemblance to nonbuilding objects. The existing GEographic Object-Based Image Analysis (GEOBIA) approaches often struggle with misclassification, mainly because they do not effectively leverage spatial information during the classification stage. Tabular Deep Learning (DL) models have emerged as promising alternatives to shallow classifiers, demonstrating improved performance in recent years. However, their effectiveness in extracting buildings based on roof color has not been thoroughly examined. To address these challenges and improve the accuracy of building extraction, this study proposed a target-based model that incorporates a sequential extraction process using tabular DL models such as Gated Additive Tree Ensemble (GATE), Neural Oblivious Decision Ensemble (NODE). The proposed method adopted a hybrid segmentation approach to effectively segment the image and extract each segment's spectral, textural, geometrical, and contextual features. Non-building objects were systematically eliminated based on their representation within the image. Subsequently, buildings with varying roof colors—brown, black, and white—were extracted. Spatial information was then meticulously integrated into the classification process. The proposed methodology was tested on a UAV image with a spatial resolution of 0.20 m and three visible bands—red, green, and blue—captured in Kingston, Ontario, Canada. This enhanced approach significantly improved precision and overall classification accuracy, demonstrating the effectiveness of incorporating spatial data into the extraction process. This research introduced a framework for incorporating spatial information into tabular models that would serve as a foundation for further improving building extraction accuracy in future studies.

Estimation of Fractal Dimension and Segmentation of Brain Tumor with Parallel Features Aggregation Network

The accurate recognition of a brain tumor (BT) is crucial for accurate diagnosis, intervention planning, and the evaluation of post-intervention outcomes. Conventional methods of manually identifying and delineating BTs are inefficient, prone to error, and time-consuming. Subjective methods for BT recognition are biased because of the diffuse and irregular nature of BTs, along with varying enhancement patterns and the coexistence of different tumor components. Hence, the development of an automated diagnostic system for BTs is vital for mitigating subjective bias and achieving speedy and effective BT segmentation. Recently developed deep learning (DL)-based methods have replaced subjective methods; however, these DL-based methods still have a low performance, showing room for improvement, and are limited to heterogeneous dataset analysis. Herein, we propose a DL-based parallel features aggregation network (PFA-Net) for the robust segmentation of three different regions in a BT scan, and we perform a heterogeneous dataset analysis to validate its generality. The parallel features aggregation (PFA) module exploits the local radiomic contextual spatial features of BTs at low, intermediate, and high levels for different types of tumors and aggregates them in a parallel fashion. To enhance the diagnostic capabilities of the proposed segmentation framework, we introduced the fractal dimension estimation into our system, seamlessly combined as an end-to-end task to gain insights into the complexity and irregularity of structures, thereby characterizing the intricate morphology of BTs. The proposed PFA-Net achieves the Dice scores (DSs) of 87.54%, 93.42%, and 91.02%, for the enhancing tumor region, whole tumor region, and tumor core region, respectively, with the multimodal brain tumor segmentation (BraTS)-2020 open database, surpassing the performance of existing state-of-the-art methods. Additionally, PFA-Net is validated with another open database of brain tumor progression and achieves a DS of 64.58% for heterogeneous dataset analysis, surpassing the performance of existing state-of-the-art methods.

SR-RDFAN-LOG: Arbitrary-scale logging image super-resolution reconstruction based on residual dense feature aggregation

In the process of oil and gas exploration and development, ultrasonic logging imaging technology visually displays the lithology, structure, fractures, pores, and other characteristics of the wellbore. This technology has been widely applied in the exploration and development of oil and gas resources. However, the low circumferential sampling rate negatively affects the imaging quality and resolution during drilling logging. To address the issues of low resolution and blurriness in ultrasonic logging images caused by these factors, as well as the limitation of existing super-resolution reconstruction algorithms that can only reconstruct fixed scales, this paper presents an Image Super-Resolution Reconstruction Network called Residual Dense Feature Aggregation (SR-RDFAN-LOG), which is specifically designed to improve ultrasonic logging image resolution. The proposed network consists of two stages, including encoding and decoding. In the encoding stage, a local feature extraction module is initially designed to integrate contextual features. Subsequently, a residual dense module is introduced for extracting deep features. Finally, a spatial attention module is constructed to focus on high-frequency details. In the decoding stage, a multilayer perceptron is used to decode the feature maps and achieve super-resolution reconstruction at arbitrary scales. The method is tested on an ultrasonic logging images test set with in-distribution scale factors of × 2 and × 4, and out-of-distribution scale factors of × 8 and × 16. When the scaling factors are × 2 and × 4, compared to the best-performing method, the peak signal-to-noise ratio (PSNR) increased by 0.4892 dB and 0.3849 dB, respectively, the structural similarity (SSIM) values increased by 0.0002 and 0.0012, respectively, and the learned perceptual image patch similarity (LPIPS) decreased by 0.0001 and 0.0008, respectively. Compared to the bicubic interpolation method, the LPIPS decreased by 0.0001 and 0.0008 at scale factors of × 8 and × 16, PSNR increased by 1.9614 dB and 2.1002 dB, respectively, and SSIM values increased by 0.0476 and 0.033, respectively. The LPIPS values decreased by 0.2233 and 0.1070, respectively. The experimental results demonstrate that our method outperforms other mainstream methods, achieves super-resolution reconstruction at arbitrary scale, and provides support for subsequent fracture-cave evaluation and oil and gas exploration.

Context-aware semantic segmentation network for tunnel face feature identification

The automated interpretation of tunnel face geological information is significant to the construction decision-making of rock mass engineering. An intelligent recognition algorithm, named the Transformer and Convolution neural networks Semantic segmentation Network (TCSeNet), is introduced to overcome the low interpretation accuracy caused by certain limitations of existing automated interpretation methods. Firstly, a hybrid encoder is constructed to extract global and local context information using Transformer and Convolution Neural Networks, respectively. Furthermore, a context feature adaptive selection module is designed to cross-fusion the information. Secondly, a feature pyramid-like decoder is constructed for multi-scale object problems. Furthermore, a multi-scale feature refinement module is designed to improve the ability of multi-scale feature expression. A tunnel face dataset is constructed for training and testing. The research shows that the mIoU of TCSeNet reaches 92.57%. This indicates that the TCSeNet is suitable for automatic interpretation of tunnel face geological information.

Exploring the Complexities of L2 English Academic Writing

L2 English academic writing, now a well-established field of study, incorporates a large number of interrelated issues from the perspectives of context, participants, process and textual realisation, informing various instructional models of English for academic purposes in university contexts globally. This paper looks at a number of major theoretical traditions and relevant research findings focusing on the characteristics of academic texts, writers, readers and the writing process, and shows how each tradition has inspired the development of particular approaches to teaching L2 English academic writing. Textual features are discussed from the points of view of genre and register analysis, and contextual features are looked at in terms of individual novice L2 writer characteristics, including writing strategies, the relationship between L1 and L2 academic writing, and the cultural background of writers. Related instructional models include product-, process-, and genre-based approaches whose operationalisations are closely linked to particular theoretical traditions. The paper argues that the pedagogical considerations stemming from different theoretical backgrounds and empirical research results can complement each other in a useful way leading to a more comprehensive pedagogical approach, and that the application of a well-informed, carefully selected and carefully sequenced combination of teaching techniques and accompanying materials can contribute to the successful development of L2 English academic writing skills.

A robust segmentation of retinal fluids from OCT images using MCFAR-net

This paper presents a novel architecture to detect Macular Edema (ME) in Optical Coherence Tomography (OCT) images that occurs due to high fluid accumulation between the retinal layers of the eye. These excessive fluids have swollen the macular region and may result in visual impairments. To alleviate this problem in its early stage, the proposed model, Multiscale Context Enhancive Aggregation & Refinement Network (MCFAR-Net) is implemented to detect these fluids enabling the prevention of visual loss. This MCFAR-Net is incorporated with two Context Feature Enhancive Modules (CFEM). The upper module is initially trained using contrastive loss for extracting the most pertinent multiscale feature maps allowing to segment thick fluid regions from OCT images. Further, the lower module elevates the feature maps to segment the minute fluid regions using the outputs of upper CFEM along with input image. Finally, the output fluid probabilistic feature maps of both paths of the two modules are stacked together and fed as input to Feature Synthesizer (FS) Module. This module improves the true positive rate of the proposed algorithm and segments the fluid region more accurately. The performance of the proposed model is trained and evaluated using publicly available datasets like RETOUCH, OPTIMA, and DUKE datasets. This model outperformed existing state-of the-art algorithms by attaining an average dice coefficient of 95.63 % when tested on the RETOUCH dataset. The performance of proposed MCFAR-Net reduced the misclassification errors enabling to identify ME more precisely in its early stage allowing the expert doctor to provide immediate treatment to the patients.

Industrial doctorates: a systematic literature review and future research agenda

ABSTRACT The aim of this systematic literature review was to understand the features, and the evolution, of industrial doctorates by analysing 54 papers published between 2002 and 2023. Two aspects were considered (i) the state of the art of the literature has been clarified through synthesis and discussion of six key themes: the main features and evolution of industrial doctorates; the design and implementation of industrial PhD programmes; the perspective of PhD candidates; the standpoint of university; the standpoint of industry; and industrial doctorates as policy tools for fostering innovation processes. There was a surge in publications after 2015, but these have mainly focused on the European context. Industrial doctorates are reshaping the doctoral education landscape as they are considered as educational tools for building bridges between academia and industry. However, there are still institutional and cultural barriers to legitimising industrial doctorates which are often considered as alternatives or even as the antithesis of ‘traditional’ PhDs; (ii) we suggest an agenda for future research to assess whether industrial doctorates do offer an effective response, or an optimal approach to meeting the demands of both the knowledge society and of doctoral candidates. Future research should seek to gather quantitative evidence in a wider variety of geographical areas, taking into account the distinctive features of both institutional contexts, and of the diverse scientific fields and industrial sectors involved. Studies of the long-term impacts of industrial doctorates on students’ research outcomes, their career trajectories and, on the performance of local ecosystems are needed.

Learning Clinical Ethics for Medical Students in Clinical Phase

Background: Clinical ethics is a method of clinical decision making ethically. The implementation of clinical ethics learning to clinical rotation students is one way to help students learn to make decisions about ethical dilemmas according to the learning outcomes in clinical rotations. There is no evaluation of the implementation of clinical ethics in clinical rotation learning in our institution, so a study is needed.Methods: This research is a descriptive study with a cross-sectional design using primary data through filling out online questionnaires (g-form) adopted from four box quadrant of clinical ethics by Johnsen, Siegler and Winslade. The respondent was clinical rotation students who underwent inpatient treatment from 10 clinical rotation stages at Zainoel Abidin Hospital. This research has been done from November until December 2022 by using consecutive sampling techniques with 75 respondent as minimum sampling. Results: A total of 87 respondents were involved in this study. This study shows that clinical ethics has been implemented in clinical rotation at Zainoel Abidin Hospital Banda Aceh. The most frequent implementation of clinical ethics is Obstetrics and Gynecology, and the least common is the Psychiatric. The implementation of clinical ethics learning in clinical rotation is already good, as seen from the majority of respondents who often get learning related to the medical indications quadrant (52.9%-58.6%), patient preference quadrant (44.8%-64.4%), the quality of life quadrant (57.5%-63.2%), and the contextual features quadrant (40.2%-59.8%).Conclusion: The implementation of clinical ethics learning in clinical rotation at Zainoel Abidin Hospital Banda Aceh is already good, although not optimal. Continuous learning of clinical ethics is necessary so that clinical rotation students are accustomed to solving ethical dilemmas in clinical situations.

Deliberative forums in fragile contexts: Challenges from the field

What does it actually take to run a deliberative mini-public (DMP)? This article unmasks the taken-for-granted assumptions of one of the most popular democratic innovations in the past decade by applying their design features in fragile political contexts. Drawing on 3 years of fieldwork in communities recovering from armed conflict and police brutality in the Philippines, we identify foundational challenges to DMPs in relation to their core design features: sortition and long-form deliberation. Our goal in documenting these challenges is not to devalue DMPs but to spotlight limitations that cannot be overcome despite the best intentions. In so doing, this article hopes to advance an honest conversation about what these DMPs can reasonably achieve, as well as the harms they can potentially cause, especially in societies where precarity is the norm rather than the exception.

Spatial-temporal graph Transformer for object tracking against noise spoofing interference

Achieving accurate object tracking against noise spoofing interference caused by similar backgrounds, similar objects, occlusion, or illumination variations is a challenge, especially since Siamese trackers largely ignore the spatial-temporal and part-level information of object structure. To handle this problem, we present a novel object tracking network with spatial-temporal awareness, which learns the correlation of local feature changes and extracts global spatial-temporal fusion features of the object to prevent noise spoofing interference. Specifically, our tracker integrates an object graph reconstruction representation module and a spatial-temporal graph Transformer module. The graph reconstruction representation module models the object structure and search region as part-to-part graph node correspondences, propagating object information to achieve part-level feature aggregation in the search region. The proposed spatial-temporal graph Transformer module fuses the temporal and spatial features of the object to enhance the contextual features and correlation between temporal and spatial feature variations, recognizing the noise spoofing interference to enhance the tracking performance. Meanwhile, our tracker learns object motion information to improve object state awareness, enhancing tracking accuracy under similar objects interference. Extensive experiments on six public datasets validate that our tracker outperforms related state-of-the-art trackers and achieves accurate tracking results against noise spoofing interference.

Induced abortion among adolescent girls and young women: should geography matter in Ghana?

Globally, there has been a slight reduction in abortion rates. However, abortion rates have increased in other major regions, including sub-Saharan Africa. Induced abortion rates vary by ecological zone and place of residence, and it is prevalent among women aged 15–24. This study examined the predictors of induced abortion in ecological zones by place of residence among adolescent girls and young women aged 15–24. The 2017 Ghana Maternal and Health Survey, which is a nationally representative household survey dataset, was used for this study. This analysis is restricted to 3194 Adolescent Girls and Young Women (AGYW) aged 15–24 years. A two-level multilevel binary logistic regression model was used to analyze the individual and contextual level characteristics associated with induced abortion. The results show that the prevalence rate of induced abortion among adolescent girls and young women is 26.0%. After adjusting for individual and community level factors, ecological zones and place of residence predicted induced abortion among AGYW. More importantly, the results demonstrate that AGYW residing in urban areas of the forest zones were statistically significant to report induced abortion compared to those in the rural areas. In conclusion, there is a strong association between place of residence, ecological zones and induced abortion in Ghana. Primarily, urban-dwelling women in the forest zone are more likely to report induced abortion. There is, therefore, a need to prioritize women residing in urban areas and the forest zone of Ghana. AGYW in these high-risk ecological zones should be reached with sufficient health education and sensitization about induced abortion. Given the protective factor of contraceptive use, it is recommended that the government and its health agencies strengthen campaigns to improve contraceptive use among AGYW across the country, particularly in the urban forest ecological zone.

Age-related contextual cueing features are more evident in reaction variability than in reaction time.

Visual-spatial contextual cueing learning underpins the daily lives of older adults, enabling them to navigate their surroundings, perform daily activities, and maintain cognitive function. While the contextual cueing effect has received increasing attention from researchers, the relationship between this cognitive ability and healthy ageing remains controversial. To investigate whether visual-spatial contextual cueing learning declines with age, we examined the contextual learning patterns of older (60-71 years old) and younger adults (18-26 years old) using a contextual-guided visual search paradigm and response variability measurements. We observed significant contextual learning effects in both age groups, impacting response speed and variability, with these effects persisting for at least 24 days. However, older adults required more repetitions and memorised fewer repeated stimuli during initial learning. Interestingly, their long-term memory maintenance appeared stronger, as their contextual facilitation persisted in both response speed and variability, while younger adults only persisted in response speed but not variability. Overall, our results suggest an age-related complex and diverse contextual cueing pattern, with older adults showing weaker learning but stronger long-term memory maintenance compared with younger adults.

Landslide extraction from aerial imagery considering context association characteristics

Accurate extraction of landslide information is crucial for timely disaster emergency response, yet this process faces significant challenges due to the interference of bare land and vegetation cover as well as the variability in landslide scales. Furthermore, the scarcity of comprehensive landslide datasets has hindered the in-depth exploration of deep learning techniques in this domain. Addressing these issues, this study introduces a novel framework for landslide extraction that leverages the characteristics of context association. In the encoding phase, a two-branch multiscale context feature extraction module (TMCFM) is established, which captures the contextual relationships across different scales through an attention mechanism while concurrently extracting context information within the same scale. To refine the interplay between features at different levels, self-attention is utilized, allowing for the effective fusion of context information at both ends of the same scale through feature fusion exploiting static and dynamic context feature modules (FSDC). Building upon this, this study develops a deeply supervised classifier (DSC) that enhances the network’s discriminative capabilities in the prediction phase via six auxiliary branches. Additionally, this study contributes a new aerial image dataset for landslide extraction created through meticulous visual interpretation. The proposed method is thoroughly compared with 17 contemporary deep learning methods, demonstrating an increase in the intersection-over-union (IoU) metric of 0.92%-16.94% over these methods. The robustness and superiority of the proposed approach are further validated through various discussions and analyses.