PrefaceList of AbbreviationsChapter 1 Introduction1��1 Background1��1��1 The 'Semantic Gap1��1��2 Query by Keywords1��2 Objectives1��3 Contributions of this Book1��3��1 Identifying Existing Semantic Learning Techniques1��3��2 Designing Effective Feature Extraction Methods for Arbitrary-Shaped Regions\"1��3��3 High-Level Concept Learning Using Decision Tree1��3��4 Applying RBIR with Semantics to Web Image Search1��4 Organization of the BookChapter 2 Key Techniques in Semantic-Based Image Retrieval2��1 Introduction2��2 Techniques and Issues in Region-Based Image Retrieval2��2��1 Image Segmentation2��2��2 Low-Level Image Feature Extraction2��2��3 Similarity Measure2��2��4 Test Database and Performance Evaluation2��3 High-Level Image Semantic Learning Techniques2��3��1 Object-Ontology2��3��2 Machine Learning2��3��3 Relevance Feedback (RF)2��3��4 Semantic Template2��3��5 Fusion of Multiple Resources for Web Image Search2��3��6 Deep Learning2��3��7 Summary of Existing Techniques in Image Semantic Learning2��4 Research Problems Addressed in this BookChapter 3 Deriving Image Semantics from Color Features3��1 Introduction3��2 Region Color Feature Extraction and Semantic Color Naming3��2��1 Region Color Features3��2��2 Semantic Color Names3��3 Image Retrieval using Semantic Color Names3��3��1 RBIR with Semantic Color Names3��3��2 Feature Normalization3��3��3 Image Similarity Measure using EMD3��4 Results and Analysis3��4��1 Test Database and Performance Evaluation Model3��4��2 Comparison of Different Color Features3��4��3 Performance of the Proposed Color Naming Method3��4��4 Image Retrieval with Color Names�� Region Color Features and GlobalColor Features3��5 Discussion and ConclusionsChapter 4 Effective Texture Feature Extraction from Arbitrary-Shaped Regions4��1 Introduction4��2 Deriving Texture Features from Arbitrary-Shaped Regions4��2��1 Projection onto Convex Set (POCS) Theory4��2��2 Extracting Region Texture Features Using POCS-ER4��2��3 Theoretical Analysis of POCS-ER4��2��4 Implementation of POCS-ER4��3 POCS-ER on Brodatz Textures4��3��1 Illustration of POCS-ER Process4��3��2 Performance of POCS-ER Measured by PSNR4��3��3 Performance of POCS-ER Measured by Retrieval Performance4��4 POCS-ER for Real-World Image Retrieval4��4��1 Experimental Setups4��4��2 Performance of Different Texture Feature Extraction Methods in RBIR������4��4��3 RBIR with Color�� Texture�� Color & Texture4��4��4 Comparison of Region Features and Global Features in Image Retrieval4��5 Conclusions and DiscussionChapter 5 Deriving High-Level Image Concepts Using Decision Tree Learning5��1 Introduction5��2 Decision Tree Learning5��2��1 Overview5��2��2 Decision Tree Induction for Image Semantic Learning5��3 The Proposed Decision Tree Induction Algorithm DT-ST5��3��1 Semantic Template Construction5��3��2 Image Feature Discretization5��3��3 Decision Tree Induction5��4 Results and Analysis5��4��1 Selection of Pre-pnming Threshold5��4��2 Pruning Unknowns5��4��3 Handling Queries with Concepts outside the Training Concept Set5��4��4 Comparison of DT-ST with ID3 and C4��55��5 Region-Based Image Retrieval with High-Level Semantics5��6 Discussion5��6��1 Scalability of DT-ST5��6��2 The Advantage of Image Retrieval with High-Level Concepts5��7 ConclusionsChapter 6 Application of Semantic-Based RBIR to Web Image Search6��1 Introduction6��2 The False Filtering Algorithm6��3 Results and Analysis6��3��1 Web Image Collection and Performance Evaluation6��3��2 Experimental Results6��4 Discussions6��4��1 Integration6��4��2 FF Response Time6��4��3 Scalability6��5 ConclusionsChapter 7 Conclusions and Future Work7��1 Conclusions of this Book7��2 Future Research DirectionsBibliographyAppendix A HSV Color Histogram and HSV-RGB ConversionAppendix B Tamura Texture FeaturesAppendix C lllustration of POCS-ER Process Using ZR and MPAppendix D Pre-pruning &Post-pruning in DT-ST