Data analytics for cultural heritage : current trends and concepts / / Abdelhak Belhi [and three others] editors

• Pubblicazione: Cham, Switzerland : , : Springer, , [2021]; ©2021
• Descrizione fisica: 1 online resource (288 pages)
• Disciplina: 363.69
• Soggetto topico: Cultural property - Data processing; Patrimoni cultural; Digitalització; Processament de dades; Aprenentatge automàtic
• Soggetto genere / forma: Llibres electrònics
• Persona (resp. second.): BelhiAbdelhak
• Nota di contenuto: Intro -- Foreword -- Preface -- Acknowledgment -- Contents -- About the Editors -- NoisyArt: Exploiting the Noisy Web for Zero-shot Classification and Artwork Instance Recognition -- 1 Introduction -- 2 Related Work -- 3 The NoisyArt Dataset -- 3.1 Data Sources -- 3.2 Data Collection -- 3.3 Discussion -- 4 Webly-Supervised Artwork Recognition -- 4.1 Baseline Classifier Model -- 4.2 Labelflip Noise -- 4.3 Entropy Scaling for Outlier Mitigation -- 4.4 Gradual Bootstrapping -- 4.5 Domain Shift Mitigation and L2 Normalization -- 5 Zero-Shot Artwork Recognition -- 5.1 Compatibility Models -- 5.2 Zero-shot Learning with Webly-Labeled Data -- 6 Experimental Results: Artwork Instance Recognition -- 6.1 Datasets -- 6.2 Webly-Supervised Classification -- 6.3 Identifying Problem Classes -- 7 Experimental Results: Zero-Shot Recognition -- 7.1 Zero-shot Recognition with Webly-Labeled Data -- 8 Conclusions and Future Work -- References -- Cultural Heritage Image Classification -- 1 Introduction -- 1.1 Artificial Neural Networks -- 2 CNN Architectures -- 3 Data and Methodology -- 3.1 Data -- 3.2 Methodology -- 3.3 Model Configuration -- 4 Results and Discussion -- References -- Study and Evaluation of Pre-trained CNN Networks for Cultural Heritage Image Classification -- 1 Introduction -- 2 Related Work -- 2.1 Feature Extraction Approaches -- 2.2 Feature Learning Approaches -- 3 The Cultural Heritage Image Classification Problem -- 3.1 Architectural Heritage Elements Dataset (AHE) -- 3.2 The WikiArt Dataset -- 4 The CNN-Based Pre-trained Networks -- 4.1 The Oxford Visual Geometry Group Models (VGG16 and VGG19) -- 4.2 Residual Networks -- 4.3 The Inception-V3 Model -- 5 Transfer Learning of the Pre-trained Networks to CH Image Classification -- 6 The Experimental Study -- 6.1 Experimental Setup -- 6.2 Experimental Results -- 6.3 Discussion -- 7 Conclusion. ; References -- Visual Classification of Intangible Cultural Heritage Images in the Mekong Delta -- 1 Background and Purpose -- 2 Approach -- 2.1 Data Collection of Intangible Cultural Heritage Images -- 2.2 Visual Approaches for Classifying Intangible Cultural Heritage Images -- 3 Results -- 3.1 Tuning Parameters -- 3.2 Classification Results for 17 ICH Categories -- 4 Conclusions -- References -- Digital Image Inpainting Techniques for Cultural Heritage Preservation and Restoration -- 1 Introduction -- 1.1 The Importance of Inpainting in Cultural Heritage -- 1.2 The Image Inpainting Problem -- 2 Interpolation -- 3 Digital Image Inpainting Methods -- 3.1 Diffusion-Based Methods -- 3.2 Texture Synthesis-Based Inpainting -- 3.3 Exemplar-Based Methods -- 3.4 Hybrid Inpainting Methods -- 3.5 Semiautomatic and Fast Inpainting Technique -- 3.6 Deep Learning-Based Technique -- 3.6.1 CNN-Based Inpainting Method -- 3.6.2 GAN-Based Inpainting Method -- 4 A Two-Stage Method for CH Digital Image Inpainting -- 5 Results and Comparisons -- 6 Conclusion -- References -- Crowd Source Framework for Indian Digital Heritage Space -- 1 Introduction -- 2 Crowd Source Framework for IHDS -- 3 Data Preprocessing -- 3.1 Redundancy Removal -- 3.2 Blur Removal -- 3.3 Super-Resolution -- 3.3.1 GAN Network Architecture -- 3.3.2 Generator Network -- 3.3.3 Discriminator Network -- 3.3.4 Perceptual Loss Function -- 4 Classification -- 4.1 Deep Neural Network -- 4.2 MobileNet Architecture -- 4.3 Transfer Learning -- 5 Results -- 5.1 IHDS Dataset -- 5.2 Blur Removal -- 5.3 Super-Resolution -- 5.3.1 Training Details -- 5.4 Transfer Learning Classification Results -- 5.5 Screenshots of Framework Working (GUI) -- 6 Conclusions -- References -- A Robust Method for Text, Line, and Word Segmentation for Historical Arabic Manuscripts -- 1 Introduction -- 2 Related Works. ; 2.1 Text Segmentation Methods -- 2.2 Line Segmentation Methods -- 2.3 Word Segmentation Methods -- 3 Proposed Method -- 3.1 Texture Component Extraction -- 3.2 Encoder-Decoder Architecture -- 3.3 Line Segmentation -- 3.3.1 Smoothing -- 3.4 Word Segmentation -- 3.4.1 The Smoothed Generalized Chamfer Distance -- 3.4.2 Classification of Inter/intra-Word Distances -- 3.5 Classification of Beginning and Ending of Words -- 3.6 Classification of First and Last Characters -- 4 Experimental Results -- 4.1 Datasets -- 4.2 Metrics -- 4.3 Analysis -- 5 Conclusion -- References -- Aesthetical Issues with Stochastic Evaluation -- 1 Introduction -- 1.1 Aesthetical Issues and Mathematics -- 1.2 Aesthetical Issues and Stochastic Analysis -- 2 Methodology -- 2.1 Stochastic Analysis in 2D -- 2.2 Illustration of Stochastic Analysis in 2D -- 3 Examples of Stochastic Analysis -- 3.1 Range of Fluctuations in Groups of Images -- 3.1.1 Data Analysis -- 3.1.2 Evaluation -- 3.2 Evaluation of Urban and Natural Landscapes Transformed by Technological and Civil Infrastructure through Climacogram Subtraction -- 3.2.1 Data Analysis -- 3.2.2 Evaluation -- 3.3 Qualitative Evaluation Aspect of Climacogram Curves -- 3.3.1 Data Analysis -- 4 Conclusions -- References -- 3D Visual Interaction for Cultural Heritage Sector -- 1 Introduction -- 2 Human-Computer Interaction -- 3 Visual Interaction in Cultural Heritage -- 4 Human-Computer Gesture Recognition -- 5 Hand Gesture Visual Processing Techniques -- 5.1 Hand Detection -- 5.1.1 3D Modelling -- 5.1.2 Pixel Values -- 5.1.3 Shape -- 5.1.4 Skin Colour -- 5.2 Hand Tracking -- 5.2.1 Estimation Based -- 5.2.2 Template Based -- 5.3 Gesture Recognition -- 6 Product Markets of Vision-Based Sensing Devices -- 6.1 Microsoft Kinect -- 6.2 Leap Motion Controller -- 7 Software Development Environments -- 7.1 Unity 3D -- 7.2 OpenCV. ; 8 Proposed Approach -- 8.1 Data Acquisition -- 8.2 Data Pre-processing -- 8.2.1 Motion Interpolation -- 8.2.2 Super-Resolution -- 8.3 Photogrammetry -- 8.4 3D Model Adaptation -- 8.5 3D Model Visualisation -- 8.6 3D Interaction -- 9 Evaluation Work and Discussion -- 9.1 Evaluation Methodology -- 9.2 Results and Discussions -- 10 Summary and Conclusions -- References -- Retrieving Visually Linked Digitized Paintings -- 1 Introduction -- 2 Related Work -- 3 Proposed Method -- 4 Experiment -- 4.1 Time Period Classification -- 4.2 Link Retrieval -- 5 Conclusion -- References -- Named Entity Recognition for Cultural Heritage Preservation -- 1 Introduction -- 1.1 Natural Language Processing in Cultural Heritage Domain -- 2 Named Entity Recognition -- 2.1 NER Process -- 2.2 NER Approaches -- 2.2.1 Rule-Based Approaches -- 2.2.2 Machine Learning Approaches -- 2.2.3 Deep Learning Approaches -- 2.3 Pre-trained NER Tools -- 2.4 Performance Measures for NER -- 3 Named Entity Recognition in Cultural Heritage Domain and Historical Texts -- 4 Discussion -- 4.1 NLP Challenges in Cultural Heritage Domain -- 4.1.1 Lack of Consistent Orthography -- 4.1.2 Solution Methods -- 5 Conclusion -- References -- Index.
• Titolo autorizzato: Data analytics for cultural heritage
• ISBN: 3-030-66777-4
• Formato: Materiale a stampa
• Livello bibliografico: Monografia
• Lingua di pubblicazione: Inglese
• Record Nr.: 9910484619103321
• Lo trovi qui: Univ. Federico II