Search Results (113)

To improve the accuracy of Japanese finger-spelled character recognition using an RGB camera, we focused on feature design and refinement of the recognition method. By leveraging angular features extracted via MediaPipe, we proposed a method that effectively captures subtle motion differences while minimizing the influence of background and surrounding individuals. We constructed a large-scale dataset that includes not only the basic 50 Japanese syllables but also those with diacritical marks, such as voiced sounds (e.g., “ga”, “za”, “da”) and semi-voiced sounds (e.g., “pa”, “pi”, “pu”), to enhance the model’s ability to recognize a wide variety of characters. In addition, the application of a change-point detection algorithm enabled accurate segmentation of sign language motion boundaries, improving word-level recognition performance. These efforts laid the foundation for a highly practical recognition system. However, several challenges remain, including the limited size and diversity of the dataset and the need for further improvements in segmentation accuracy. Future work will focus on enhancing the model’s generalizability by collecting more diverse data from a broader range of participants and incorporating segmentation methods that consider contextual information. Ultimately, the outcomes of this research should contribute to the development of educational support tools and sign language interpretation systems aimed at real-world applications. Full article

Brazil is increasingly affected by extreme weather events due to climate change, with pronounced regional differences in temperature and precipitation patterns. The southeast region is particularly vulnerable, frequently experiencing severe droughts and extreme heatwaves linked to atmospheric blocking events and intense rainfall episodes driven by the South Atlantic Convergence Zone (SACZ). These phenomena contribute to recurring climate-related disasters. The country’s heavy reliance on hydropower heightens its susceptibility to droughts, while growing evidence points to intensifying dry spells and wildfires across multiple regions, threatening agricultural output and food security. Urban areas, particularly, are experiencing more frequent and severe heatwaves, posing serious health risks to vulnerable populations. This study investigates the links between global teleconnection indices and synoptic-scale systems, specifically blocking events and SACZ activity, and their influence on Brazil’s extreme heat, drought conditions, and river flow variability over the past 30 to 40 years. By clarifying these interactions, the research aims to enhance understanding of how large-scale atmospheric dynamics shape climate extremes and to assess their broader implications for water resource management, energy production, and regional climate variability. Full article

Past research has extensively explored reading in English-speaking children with dyslexia who acquire a highly irregular and opaque orthography. Far less is known about the manifestation of dyslexia in shallow, highly consistent orthographies like German, especially in adults. To shed further light on the heterogenous manifestation of dyslexia in German-speaking adults, we assessed reading and reading-related abilities, spelling, cognitive abilities, and language learning experience in 33 healthy German-speaking adults (17 females) and 33 adults with dyslexia (20 females). The four main aims were to (1) elucidate the intricate relationship between cognitive and literacy abilities, (2) investigate persisting weaknesses, (3) determine the strongest predictors of dyslexia, and (4) investigate deficit profiles. Group comparisons revealed persistent deficits in almost all measures of reading and spelling, slight deficits in verbal working memory, but no visuospatial impairments in adults with dyslexia. Moreover, adults with dyslexia had considerably lower English skills and lower educational attainment. Overall, we found fewer and weaker links between literacy and cognitive measures in adults with dyslexia, indicating a dissociation between these skills. Spelling, word reading, and phonological awareness were the best predictors of dyslexia, but the most widespread deficit was rapid automatized naming. Our findings suggest a heterogeneous manifestation of dyslexia in German-speaking adults, with even low-level deficits persisting into adulthood despite the shallow nature of the German orthographic system. Full article

The Askisi-Spelling Deficits (SD) neuropsychological web-based screener was developed to assess cognitive and spelling abilities in children, with an emphasis on the early detection of spelling disorders. This tool incorporates six tasks that evaluate cognitive domains, such as visual and auditory working memory, response inhibition, and spelling processing, providing a comprehensive framework for assessment. A study conducted with 264 Greek children, including 132 children with spelling deficits and 132 typically developing controls, aimed to implement this screening tool. Results indicated that the screener was effective, as children with spelling deficits showed significantly lower performance and longer response times across all tasks. The tool’s internal consistency was supported by split-half correlations (r = 0.64) and Spearman–Brown coefficients (r = 0.78). Nonetheless, certain limitations were identified, including the absence of latency data for specific tasks (Go/No-Go and working memory), as well as the screener’s cultural specificity, which might limit its applicability to other linguistic and orthographic systems. Future iterations should prioritize the inclusion of timing mechanisms for more detailed assessments and consider adaptations for use in languages with varying orthographic complexities. Expanding the demographic reach and conducting longitudinal validation studies would further improve its utility and generalizability. The web-based nature of the screener enables scalable and standardized administration, making it a practical and efficient tool for the early identification of spelling difficulties in children. Full article

Brain–computer interfaces (BCIs) provide a direct communication pathway between the central nervous system and external environments, enabling human–machine interaction control. Among them, event-related potential (ERP)-based BCIs are among the most accurate and reliable BCI systems. However, current mainstream classification algorithms struggle to eliminate calibration requirements and rely heavily on costly labeled data, limiting the practical usability of ERP-based BCIs. To address this, the development of unsupervised algorithms is critical for advancing real-world BCI applications. In this study, we propose the spatio-temporal equalization sliding-window distribution distance maximization (STE-sDDM) algorithm, which introduces spatio-temporal equalization (STE) to unsupervised ERP classification for the first time and integrates it with a novel unsupervised classification method, sliding-window distribution distance maximization (sDDM). STE estimates and removes colored noise interference in background noise to enhance the signal-to-noise ratio of inputs for sDDM. Meanwhile, sDDM leverages an enhanced inter-class divergence metric based on the ergodic hypothesis theory, utilizing sliding windows to emphasize temporally discriminative features, thereby improving unsupervised classification accuracy. The experimental results demonstrate that the integration of STE and sDDM significantly enhances ERP feature separability, outperforming state-of-the-art unsupervised online classification algorithms in spelling accuracy and the information transfer rate (ITR), facilitating more accurate and faster plug-and-play real-time control for BCI systems. Additionally, static spatio-temporal equalizer architectures were found to outperform dynamic architectures when combined with this framework. Full article

Communication barriers pose significant challenges for the Deaf and Hard-of-Hearing (DHH) community, limiting their access to essential services, social interactions, and professional opportunities. To bridge this gap, assistive technologies leveraging artificial intelligence (AI) and deep learning have gained prominence. This study presents a real-time American Sign Language (ASL) interpretation system that integrates deep learning with keypoint tracking to enhance accessibility and foster inclusivity. By combining the YOLOv11 model for gesture recognition with MediaPipe for precise hand tracking, the system achieves high accuracy in identifying ASL alphabet letters in real time. The proposed approach addresses challenges such as gesture ambiguity, environmental variations, and computational efficiency. Additionally, this system enables users to spell out names and locations, further improving its practical applications. Experimental results demonstrate that the model attains a mean Average Precision (mAP@0.5) of 98.2%, with an inference speed optimized for real-world deployment. This research underscores the critical role of AI-driven assistive technologies in empowering the DHH community by enabling seamless communication and interaction. Full article

Drought is one of the most widespread natural disasters globally, and its spatiotemporal distribution is profoundly influenced by the El Niño-Southern Oscillation (ENSO). As a typical humid coastal basin, the Jiaojiang River Basin in southeastern China frequently experiences hydrological extremes such as dry spells during flood seasons. This study focuses on the Jiaojiang River Basin, aiming to investigate the response mechanisms of drought evolution to ENSO in coastal regions. This study employs 10-day scale data from 1991 to 2020 to investigate the drought mechanisms driven by ENSO through a comprehensive framework that combines standardized indices with climate–drought correlation analysis. The results indicate that the Comprehensive Drought Index (CDI), integrating the advantages of the Standardized Precipitation Index (SPI) and Standardized Runoff Index (SRI), effectively reflects the basin’s combined meteorological and hydrological wet-dry characteristics. A strong response relationship exists between drought indices in the Jiaojiang River Basin and ENSO events. Drought characteristics in the basin vary significantly during different ENSO phases. The findings can provide theoretical support for the construction of resilient regional water resource systems, and the research framework holds reference value for sustainable development practices in similar coastal regions globally. Full article

Kreol Morisien is a vibrant and expressive language that reflects the multicultural heritage of Mauritius. There are different versions of Kreol languages. While Kreol Morisien is spoken in Mauritius, Kreol Rodrige is spoken only in Rodrigues, and they are distinct languages. Being spoken by only about 1.5 million speakers in the world, Kreol Morisien falls in the category of under-resourced languages. Initially, Kreol Morisien lacked a formalised writing system, with many people using different spellings for the same words. The first step towards standardisation of writing Kreol Morisien was after the publication of the Kreol Morisien orthography in 2011 and Kreol Morisien grammar in 2012 by the Kreol Morisien Academy. Kreol Morisien obtained a national position in the year 2012 when it was introduced in educational organisations. This was a major breakthrough for Kreol Morisien to be recognised as a national language on the same level as English, French, and other oriental languages. By providing a means for Kreol Morisien speakers to connect with others, a translation system will help to preserve and strengthen the identity of the language and its speakers in an increasingly globalized world. The aim of this paper is to develop a translation system for Kreol Morisien and English. Thus, a dataset consisting of 50,000 parallel Kreol Morisien and English sentences was created, where 48,000 sentence pairs were used to train the models, while 1000 sentences were used for evaluation and another 1000 sentences were used for testing. Several machine translation systems such as statistical machine translation, open-source neural machine translation, a Transformer model with attention mechanism, and Marian machine translation are trained and evaluated. Our best model, using MarianMT, achieved a BLEU score of 0.62 for the translation of English to Kreol Morisien and a BLEU score of 0.58 for the translation of Kreol Morisien into English. To our knowledge, these are the highest BLEU scores that are available in the literature for this language pair. A high-quality translation tool for Kreol Morisien will facilitate its integration into digital platforms. This will make previously inaccessible knowledge more accessible, as the information can now be translated into the mother tongue of most Mauritians with reasonable accuracy. Full article

Thisstudy introduces a new model for detecting insults in Roman Urdu, filling an important gap in natural language processing (NLP) for low-resource languages. The transliterated nature of Roman Urdu also poses specific challenges from a computational linguistics perspective, including non-standardized grammar, variation in spellings for the same word, and high levels of code-mixing with English, which together make automated insult detection for Roman Urdu a highly complex problem. To address these problems, we created a large-scale dataset with 46,045 labeled comments from social media websites such as Twitter, Facebook, and YouTube. This is the first dataset for insult detection for Roman Urdu that was created and annotated with insulting and non-insulting content. Advanced preprocessing methods such as text cleaning, text normalization, and tokenization are used in the study, as well as feature extraction using TF–IDF through unigram (Uni), bigram (Bi), trigram (Tri), and their unions: Uni+Bi+Trigram. We compared ten machine learning algorithms (logistic regression, support vector machines, random forest, gradient boosting, AdaBoost, and XGBoost) and three deep learning topologies (CNN, LSTM, and Bi-LSTM). Different models were compared, and ensemble ones were proven to give the highest F1-scores, reaching 97.79%, 97.78%, and 95.25%, respectively, for AdaBoost, decision tree, TF–IDF, and Uni+Bi+Trigram configurations. Deeper learning models also performed on par, with CNN achieving an F1-score of 97.01%. Overall, the results highlight the utility of n-gram features and the combination of robust classifiers in detecting insults. This study makes strides in improving NLP for Roman Urdu, yet further research has established the foundation of pre-trained transformers and hybrid approaches; this could overcome existing systems and platform limitations. This study has conscious implications, mainly on the construction of automated moderation tools to achieve safer online spaces, especially for South Asian social media websites. Full article

Examining spelling from a developmental perspective began in the 1970s and has broadened over the years. This research has informed understanding of the nature and development of spelling or orthographic knowledge in children and older students and the role of orthographic knowledge in reading and writing. Based on analyses of the errors that students make in their writing and on spelling assessments, developmental spelling has documented the acquisition and integration of progressively more complex spelling patterns that represent both sound and meaning and illuminated how this information supports students’ ability to read as well as to write words. Intended for researchers, teacher educators, and teachers of students in grades 3–12, this article describes the layers of the spelling system that developmental spelling research has investigated, and their progressive integration in learners, including those who struggle, from the intermediate through the middle and secondary grades. It addresses the implications of developmental spelling research for assessment and instruction in spelling, word analysis, vocabulary, and the more specific implications of developmental spelling research for aligning instruction across spelling, word analysis, vocabulary, morphology, and etymology. Full article

P300 detection is a difficult task in brain–computer interface (BCI) systems due to the low signal-to-noise ratio (SNR). In BCI systems, P300 waves are generated in electroencephalogram (EEG) signals using various oddball paradigms. Convolutional neural networks (CNNs) have previously shown excellent results for P300 detection compared to different machine learning models. However, current CNN architectures limit P300 detection accuracy because these models usually only extract single-scale features. Aiming to enhance P300 detection accuracy, an inception module-based CNN architecture, namely Inception-CNN, is introduced. Inception-CNN effectively learns discriminative features from both spatial and temporal information to reduce overfitting and computational complexity. Furthermore, it can extract multi-scale features, which effectively improves P300 detection accuracy and increases character spelling accuracy. To analyze the effect of the inception layer, two additional models are proposed: Inception-CNN-S, which uses the inception layer with a spatial convolution layer, and Inception-CNN-T, which uses the inception layer with a temporal convolution layer. The proposed model was evaluated on dataset II of BCI Competition III and dataset IIb of BCI Competition II. The experimental results show that Inception-CNN provides a promising solution for improving the accuracy of P300 detection, with F1 scores of 47.14%, 55.28%, and 78.94% for dataset II of BCI Competition III (Subject A and Subject B) and dataset IIb of BCI Competition II, respectively. Full article

The Mexican deaf community primarily uses Mexican Sign Language (MSL) for communication, but significant barriers arise when interacting with hearing individuals unfamiliar with the language. Learning MSL requires a substantial commitment of at least 18 months, which is often impractical for many hearing people. To address this gap, we present an MSL-to-Spanish translation system that facilitates communication through a spelling-based approach, enabling deaf individuals to convey any idea while simplifying the AI’s task by limiting the number of signs to be recognized. Unlike previous systems that focus exclusively on static signs for individual letters, our solution incorporates dynamic signs, such as “k”, “rr”, and “ll”, to better capture the nuances of MSL and enhance expressiveness. The proposed Hybrid Neural Network-based algorithm integrates these dynamic elements effectively, achieving an F1 score of 90.91%, precision of 91.25%, recall of 91.05%, and accuracy of 91.09% in the extended alphabet classification. These results demonstrate the system’s potential to improve accessibility and inclusivity for the Mexican deaf community. Full article

This study examines the combined effects of decommissioning lignite mining operations and long-term climate trends on groundwater systems and land surface movements in the Konin region of Poland, which is characterised by extensive open-pit lignite extraction. The findings reveal subsidence rates ranging from −26 to 14 mm per year within mining zones, while land uplift of a few millimetres per year occurred in closed mining areas between 2015 and 2022. Groundwater levels in shallow Quaternary and deeper Paleogene–Neogene aquifers have declined significantly, with drops of up to 26 m observed near active mining, particularly between 2009 and 2019. A smaller groundwater decline of around a few metres was observed outside areas influenced by mining. Meteorological data show an average annual temperature of 8.9 °C from 1991 to 2023, with a clear warming trend of 0.0050 °C per year since 2009. Although precipitation patterns show a slight increase from 512 mm to 520 mm, a shift towards drier conditions has emerged since 2009, characterised by more frequent dry spells. These climatic trends, combined with mining activities, highlight the need for adaptive groundwater management strategies. Future research should focus on enhanced monitoring of groundwater recovery and sustainable practices in post-mining landscapes. Full article

Named entity recognition is a critical task in the electronic medical record management system for rehabilitation robots. Handwritten documents often contain spelling errors and illegible handwriting, and healthcare professionals frequently use different terminologies. These issues adversely affect the robot’s judgment and precise operations. Additionally, the same entity can have different meanings in various contexts, leading to category inconsistencies, which further increase the system’s complexity. To address these challenges, a novel medical entity recognition algorithm for Chinese electronic medical records is developed to enhance the processing and understanding capabilities of rehabilitation robots for patient data. This algorithm is based on a fusion classification strategy. Specifically, a preprocessing strategy is proposed according to clinical medical knowledge, which includes redefining entities, removing outliers, and eliminating invalid characters. Subsequently, a medical entity recognition model is developed to identify Chinese electronic medical records, thereby enhancing the data analysis capabilities of rehabilitation robots. To extract semantic information, the ALBERT network is utilized, and BILSTM and MHA networks are combined to capture the dependency relationships between words, overcoming the problem of different meanings for the same entity in different contexts. The CRF network is employed to determine the boundaries of different entities. The research results indicate that the proposed model significantly enhances the recognition accuracy of electronic medical texts by rehabilitation robots, particularly in accurately identifying entities and handling terminology diversity and contextual differences. This model effectively addresses the key challenges faced by rehabilitation robots in processing Chinese electronic medical texts, and holds important theoretical and practical value. Full article

When using runoff infiltration devices to remove nitrogen and phosphorus pollutants from urban runoff, the quality of the effluent is affected by the length of dry spells between rain events. This study presents a novel analysis of how these dry periods impact the device’s effectiveness in removing pollutants and the resulting biological succession within the filter. Our analysis examines nitrogen and phosphorus removal in a rainwater filtration context, providing new insights into how dry period duration influences infiltration system performance. The results indicate that biological processes have a significant impact on reducing total nitrogen (TN) and total phosphorus (TP) contents under different drying periods. A 3-day drying period is most effective for reducing TN through biological processes, while a 7-day period is best for TP reduction. This suggests that moderately extending the drying period improves TP removal efficiency but does not enhance TN removal. The dominant bacterial phylum responsible for denitrification and phosphorus removal is Proteobacteria, with Pseudomonas and Acinetobacter as the leading genera. As the drying period lengthens, the dominant genera shift from Pseudomonas to Massilia. At a 3-day drying period, denitrification primarily occurs through Pseudomonas on the surfaces of maifanite and zeolite. At a 7-day dry-out period, Acinetobacter is mainly responsible for phosphate removal on maifanite surfaces. However, after a 14-day dry-out period, both biomass and bioactivity of Pseudomonas and Acinetobacter decrease, leading to reduced efficiency in removing nitrogen and phosphorus pollutants from runoff infiltration devices. These results aid in developing runoff infiltration devices for specific scenarios and offer crucial guidance for regulating runoff pollution control technologies. Full article