Advances in pattern recognition research

Intro; Contents; Preface; Chapter 1; Automatic Target Recognition Processor Using Integrated Grayscale Optical Correlator and Neural Network; Abstract; 1. Introduction; 2. Grayscale Optical Correlator for Target Detection; 2.1. Grayscale Optical Correlator System Space-Bandwidth-Product Matching; 2....

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Weitere Verfasser:	Lu, Thomas (HerausgeberIn), Chao, Tien-Hsin (HerausgeberIn)
Format:	UnknownFormat
Sprache:	eng
Veröffentlicht:	New York Nova Science Publishers 2018
Schriftenreihe:	Computer science, technology and applications
Schlagworte:	Pattern recognition systems > Pattern perception > Research > Machine learning > Mustererkennung > Maschinelles Lernen
Online Zugang:	Inhaltsverzeichnis
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Bestellen

Beschreibung
Zusammenfassung:	Intro; Contents; Preface; Chapter 1; Automatic Target Recognition Processor Using Integrated Grayscale Optical Correlator and Neural Network; Abstract; 1. Introduction; 2. Grayscale Optical Correlator for Target Detection; 2.1. Grayscale Optical Correlator System Space-Bandwidth-Product Matching; 2.2. Input SLM Selection; 3. Miniaturized Grayscale Optical Correlator; 3.1. 512 x 512 GOC System Architecture; 3.2. Graphic User Interface of the GOC System; 3.3. Grayscale Optical Correlator Testing; 3.4. Summary; 4. Composite OT-MACH Correlation Filter; 4.1. Principle of OT-MACH Filter. 4.2. Automatic Optimization of OT-MACH Filter4.2.1. Optimization Approach; 4.2.2. Test Results; 4.2.3. Discussion and Summary; 5. Optically Implementation of OT-MACH Filter; 5.1. Filter Projection to Dynamic-Range-Limited Real-Valued SLM; 5.2. Summary; 6. Second Stage: Neural Network Target Recognition and Classification; 6.1. Sonar Mine ATR Processing -- An Illustrating Example; 6.2. Wavelet Transform; 6.3. Feature Extraction; 6.4. Neural Network; 6.5. Results and Discussion; Conclusion; Acknowledgments; References; Chapter 2; Deep Neural Networks for Pattern Recognition; Abstract. 1. Pattern Recognition in Human Vision2. Human Vision-Inspired Conditional Generative Adversarial Networks; 3. Precision Multi-Band Infrared Image Segmentation Using Conditional Generative Adversarial Networks; 4. Occluded Object Reconstruction Using Conditional Generative Adversarial Networks; 5. Image Enhancement from Visual to Infrared Using Conditional Generative Adversarial Networks; 6. Data Augmentation for Training Deep Neural Networks; 7. Incremental Training; Conclusion; Acknowledgments; References; Chapter 3; Robust Pattern Recognition via Joint Transform Correlation; Abstract. 1. Introduction2. Theoretical Analysis; 2.1. Fringe-Adjusted Joint Transform Correlation (FJTC); 2.2. Logarithmic FJTC (LFJTC); 2.3. Shifted Phase-Encoded Fringe-Adjusted JTC (SPJTC); 2.4. Gaussian Filter Based SPJTC (G-SPJTC); 2.5. Gaussian Filter Based LFJTC (G-LFJTC); 3. Experimental Results; 3.1. Dataset Description; 3.2. Results and Comparison; Conclusion; References; Chapter 4; The Spatial Domain Optimal Trade-Off Maximum Average Correlation Height Filter and Its Performance Assessment; Abstract; 1. Introduction; 2. Frequency Domain Design of the OT-MACH Filter. 3. Performance Limitations of the Frequency Domain OT-MACH Filter4. Spatial domain Optimal Trade-Off Maximum Average Correlation Height (SPOT-MACH) Filter; 4.1. Why Spatial Domain Correlation Filters?; 4.2. Design of the SPOT-MACH Filter; 4.3. SPOT-MACH Filter Implementation; 4.4. Operation of the SPOT-MACH Filter; 4.5. Performance Assessment of the SPOT-MACH Filter; 5. Performance of the SPOT-MACH Filter with Infra-Red Imagery; 6. Comparison of the SPOT-MACH Filter and the Shift Invariant Feature Transform Using FLIR Imagery; Conclusion; References; Chapter 5.
Beschreibung:	Includes bibliographical references and index
Beschreibung:	xiii, 272 Seiten Illustrationen, Diagramme 24 cm
ISBN:	1536144290 1-5361-4429-0 9781536144291 978-1-5361-4429-1