Band 54, Heft 4

This study presents an integrated and extensible OCR framework that transforms physical documents into searchable digital archives, emphasizing system-level engineering over algorithmic novelty. While the core components such as contrast enhancement, PaddleOCR, and BERT-based spelling correction are individually established, their synthesis into a unified pipeline for multilingual and multimodal archival digitization represents a practical contribution to digital preservation. The architecture is implemented using Python and incorporates preprocessing steps such as grayscale conversion, CLAHE-based contrast adjustment, Gaussian blurring, and adaptive thresholding to enhance image quality. Multilingual high-accuracy text recognition is performed using PaddleOCR, while a hybrid spelling correction module combines lexical and BERT-based contextual corrections. A straightforward and easy-to-use graphical user interface (GUI) facilitates interaction and visual diagnostics on the basis of confidence heatmaps as well as pixel intensity histograms. Digitized outputs are retained in a central library repository to facilitate efficient retrieval and organization. The system performs well in high OCR accuracy across different input types 99.12 % in screenshots and 94.81 % in scanned documents. Preprocessing greatly enhances text readability in degraded images. The spelling correction module enhances text readability by more than 95 %. Visualization tools offer useful insights into OCR performance and preprocessing effects. This paper demonstrates a modular pipeline that integrates preprocessing, OCR, error correction, visualization, and repository integration. The application of PaddleOCR with domain-specific preprocessing and contextual correction capabilities brings novelty, particularly for rich archival content.

Three-dimensional (3D) modelling is an efficient technique for preserving visual archaeological legacy by offering chronological evidence of architectural alterations. This research used a mixed-methods approach to explore the Itakhola Mound Temple, particularly its five phases, as a notable archaeological site in the Lalmai-Mainamati Hill region of the Cumilla district, Bangladesh. All the available evidence and field observations suggest that the highest reconstruction phase is situated above the other four phases. Consequently, only the fifth phase is observable, as opposed to the others. To render all phases visibly, the solid 3D model has been generated utilizing several software tools, for example, Autodesk AutoCAD, 3D Studio Max, and Adobe Photoshop. These elucidated the chronological architectural modifications of the temple and the likely reasons for these alterations. This research elucidates the rationale for choosing solid modelling over photogrammetry or laser scanning methods, as well as the alterations in measurement over several phases. This study emphasizes the significance of virtual preservation, specifically concerning temple architecture only, rather than environmental elements. This demonstrates that 3D solid modelling functions as a sustainable digital approach for the virtual preservation and representation of archaeological monuments, thereby aiding future investigations and conservation efforts.

Online Finding Aids (OFA) portals are essential for providing access to archival records. However, many existing systems suffer from fragmented interfaces that separate content from catalogue access, leading to poor usability and user dissatisfaction. This study aims to design site navigation and wireframes for an integrated OFA portal that enhances user experience, accessibility, and usability through a unified, single-window approach. The research combines content analysis and design thinking to address this challenge. First, 10 prominent archival websites were evaluated to identify patterns and issues in navigation and content presentation. These findings informed a structured design thinking process comprising empathy mapping, problem definition, ideation, prototyping, and testing conducted through collaborative workshops with archivists, researchers, and designers. Wireframes were developed using Figma to visually represent an integrated platform that unifies content management and archival catalogues within a streamlined interface. Unlike traditional archival systems, which require users to switch between separate websites for institutional content and finding aids, the proposed wireframe consolidates both into a single, user-tested platform. Usability testing with 10 participants using the System Usability Scale (SUS) yielded an average score of 84, indicating excellent usability and user satisfaction. The resulting wireframe introduces intuitive navigation and integrated content management features that eliminate the need for separate systems. This study contributes a replicable and evidence-based framework for improving the usability of archival websites and sets a practical foundation for developing user-centred online finding aids.

The preservation of paper-based cultural heritage materials is a persistent challenge, especially in tropical and resource-limited contexts where environmental and biological stressors accelerate deterioration. This study investigates the application of deep learning to detect early-stage biodeterioration – specifically insect and mold damage – on digitized paper artefacts. A curated dataset of 274 conservation-grade images was annotated into two classes and evaluated using stratified three-fold cross-validation. Four ImageNet-pretrained convolutional neural networks (CNNs) – VGG-16, ResNet-50, ResNet-101, and MobileNet V2 – were benchmarked. While all models achieved comparable mean accuracy (∼0.719), MobileNet V2 delivered a higher weighted F 1 -score (0.681) and significantly lower training latency outperforming deeper architectures in both efficiency and minority-class detection. This result is particularly relevant for preventive conservation workflows that require rapid identification of rare deterioration events. MobileNet V2’s lightweight architecture also enables potential deployment on edge devices, supporting real-time diagnostics in collection environments with limited computing infrastructure. The findings affirm the viability of compact CNNs for accurate, class-balanced performance under practical constraints. Limitations include a relatively small dataset and controlled imaging conditions, suggesting the need for expanded, multi-institutional corpora and more diverse acquisition scenarios. Future work will incorporate model explainability techniques (e.g., Grad-CAM) and explore transformer-based alternatives. This research contributes a scalable, reproducible AI-based framework for integrating machine learning into digital heritage diagnostics, advancing the intersection of conservation science and library technology.

Based on knowledge visualization tools such as CiteSpace and VOSviewer, this study conducts a systematic review and visual analysis of literature on the Chinese intangible cultural heritage (ICH) tradition known as Taige, covering the period from 2000 to 2024. Using multiple dimensions, including keyword co-occurrence, burst detection, clustering structure, and temporal evolution, the paper reveals the spatial-temporal distribution, research hotspots, and emerging trends in the field. The findings suggest that Taige, a significant form of China’s ICH, has shifted from early emphases on performance documentation and folkloric description toward interdisciplinary explorations that integrate heritage preservation, digital transformation, design empowerment, and cultural communication. Keyword clustering and burst analysis highlight a growing convergence of multidisciplinary approaches, the integration of theory and practice, and a deep coupling of “Culture + Design.” Looking ahead, the protection and dissemination of Taige should prioritize the application of visual design, digital interaction, and immersive technologies to promote its creative transformation and design-oriented transmission. This study provides both theoretical insight and data visualization for the design-centered ICH research, offering a referential framework for future inquiry.

Digital preservation is crucial for ensuring the long-term accessibility and usability of digital content; however, many institutions struggle with inadequate staffing, policies, and financial resources. While previous studies have primarily focused on the technical skills required for digital preservation, this study takes a novel approach by assessing the leadership and advocacy capabilities crucial for initiating and sustaining digital preservation programs. Data was collected from 178 librarians using a pilot-tested, structured questionnaire. Librarians demonstrated moderate competencies in budget management, staff motivation, and collaboration, while policy development and legal/copyright issues were identified as weaker areas. Significant differences were observed based on institution type, with degree-awarding institutions scoring higher in financial planning and policy formulation. Librarians from better-resourced regions exhibited stronger capabilities than those from underserved areas. Higher academic qualifications and greater professional experience correlated with stronger competencies in leadership, policymaking, and securing funding. The study highlights the need for targeted training programs, mentorship initiatives, and structured policies to enhance librarians’ strategic and managerial competencies in digital preservation. The study underscores the crucial role of leadership and advocacy in aligning institutional strategies with digital preservation goals.

Preprints have become an increasingly important component of scholarly communication, offering rapid dissemination of research prior to formal peer review. This study presents a comprehensive bibliometric analysis of peer-reviewed, Scopus-indexed articles published between 2015 and 2024 that examine preprints, preprint platforms, and their relationship to open science. Using Bibliometrix and VOSviewer, this study analyzes 4,826 articles to identify publication trends, prolific authors, influential journals, collaboration patterns, and thematic research clusters. Results reveal a steady increase in scholarly attention toward preprints, with strong linkages to topics such as open access, research transparency, and rapid knowledge sharing. Thematic mapping highlights core research hotspots in policy development, platform infrastructure, and disciplinary differences in adoption, alongside emerging themes such as preprints in health crises and their role in equitable knowledge dissemination. The analysis also uncovers gaps in cross-disciplinary collaboration and qualitative assessments of preprint impact. Future trends point toward broader policy alignment, enhanced interdisciplinary networks, and deeper integration with open science frameworks. To accelerate adoption, stakeholders in related fields can promote preprints by raising institutional awareness, providing training on platform use, and encouraging journal and funder policies that recognize preprint contributions.