Advanced Concepts for Intelligent Vision Systems 17th International Conference, ACIVS 2016, Lecce, Italy, October 24-27, 2016, Proceedings

• Main Authors: Blanc-Talon, Jacques, Distante, Cosimo, Philips, Wilfried, Popescu, Dan, Scheunders, Paul
• Format: eBook
• Language: English
• Published: Netherlands Springer Nature 2016; Springer International Publishing AG; Springer
• Edition: 1
• Series: LNCS sublibrary. SL 6, Image processing, computer vision, pattern recognition, and graphics
• Subjects: Computer programming, programs, data; Computer vision-Congresses
• ISBN: 3319486802; 9783319486802; 9783319486796; 3319486799

This book constitutes the refereed proceedings of the 17th International Conference on Advanced Concepts for Intelligent Vision Systems, ACIVS 2016, held in Lecce, Italy, in October 2016.The 64 revised full papers presented in this volume were carefully selected from 137 submissions. They deal with ...

Table of Contents:

• Intro -- Preface -- Organization -- Contents -- Gradients versus Grey Values for Sparse Image Reconstruction and Inpainting-Based Compression -- 1 Introduction -- 2 PDE-Based Reconstructions from Sparse Data -- 2.1 Reconstructions from Grey Values -- 2.2 Reconstructions from Gradients -- 3 Data Optimisation for PDE-Based Reconstructions -- 3.1 Optimisation Algorithms for Grey Values -- 3.2 Optimisation Algorithms for Gradients -- 4 Encoding Framework -- 4.1 Encoding of Grey Value Data -- 4.2 Encoding of Gradient Data -- 5 Evaluation of Gradients Versus Grey Values -- 5.1 PDE-Based Image Reconstruction -- 5.2 PDE-Based Image Compression -- 6 Conclusions -- References -- Global Bilateral Symmetry Detection Using Multiscale Mirror Histograms -- 1 Introduction -- 2 Related Work -- 3 Algorithm Details -- 3.1 Local Orientation Histogram -- 3.2 Multi-scale Information -- 3.3 Pairwise Symmetry Coefficient and Mirror Symmetry Histogram -- 4 Experiments and Discussions -- 4.1 Implementation and Evaluation Details -- 4.2 Comparison on PSU Datasets (2010, 2011 and 2013) -- 4.3 Comparison on AVA Dataset -- 5 Conclusion -- References -- Neural Network Boundary Detection for 3D Vessel Segmentation -- 1 Introduction -- 2 Method -- 2.1 Initial Segmentation -- 2.2 Neural Network Boundary Detection -- 2.3 Segmentation with B-Spline Mesh Regularisation -- 3 Application and Results -- 3.1 Boundary Detection Results -- 3.2 Segmentation Results -- 4 Conclusion -- References -- A Simple Human Activity Recognition Technique Using DCT -- Abstract -- 1 Introduction -- 2 Discrete Cosine Transform (DCT) -- 3 DCT Based Human Activity Recognition Method -- 4 Experimental Results -- 5 Conclusion -- References -- Hand Gesture Recognition Using Infrared Imagery Provided by Leap Motion Controller -- 1 Introduction -- 2 DSQP-Based Feature Description -- 3 Compressive Sensing
• 5.2 Detection -- 5.3 Compressed Models -- 6 Conclusions -- References -- Factor Analysis of Dynamic Sequence with Spatial Prior for 2D Cardiac Spect Sequences Analysis -- 1 Introduction -- 2 Methods -- 3 Algorithm and Implementation Details -- 4 Results -- 4.1 Synthetic Data -- 4.2 6-DIG SPECT Sequence -- 5 Conclusion -- References -- Soccer Player Detection with only Color Features Selected Using Informed Haar-like Features -- 1 Introduction -- 2 Informed Haar-like Features for Human Detection -- 3 Soccer Player Detection with only Color Features Selected Using Informed Haar-like Features -- 3.1 Dataset Used in Our Method -- 3.2 Design of Templates -- 3.3 Structure of Weak Classifiers -- 3.4 Update of Training Samples and Construction of a Strong Classifier -- 4 Evaluation -- 4.1 Image Resolution and Window Size for Detection -- 4.2 Comparison About Color Spaces -- 4.3 Influence of the Number of Training Iterations and Depth of the Weak Classifier on Accuracy -- 4.4 Processing Speed -- 4.5 Detection Examples and Discussion -- 5 Conclusion -- References -- Person Re-identification in Frontal Gait Sequences via Histogram of Optic Flow Energy Image -- 1 Introduction -- 2 Related Work -- 3 Methodology -- 4 Experimental Results -- 4.1 Re-ID in Controlled Scenario: CASIA Dataset -- 4.2 Re-ID in Uncontrolled Scenario: HDA Person Dataset -- 5 Conclusions and Future Work -- References -- A Bayesian Approach to Linear Unmixing in the Presence of Highly Mixed Spectra -- 1 Introduction -- 2 Mathematical Model -- 2.1 Linear Mixing Model -- 2.2 Bayesian Formulation -- 2.3 Gibbs Sampling Algorithm -- 2.4 Simulated Annealing for Bounds Estimation -- 3 Results on Synthetic and Real Data -- 4 Conclusion -- References -- Key Frames Extraction Based on Local Features for Efficient Video Summarization -- Abstract -- 1 Introduction -- 2 Related Works
• 4 Classification Process -- 5 Leap Motion Controller -- 6 Results -- 7 Conclusions -- References -- Horizon Line Detection from Fisheye Images Using Color Local Image Region Descriptors and Bhattacharyya Coefficient-Based Distance -- 1 Introduction -- 2 Image Simplification -- 2.1 Colorimetric Invariants (CI) -- 2.2 Exponential Transform (ET) -- 3 Initial Segmentation Using Statistical Region Merging -- 4 Region Feature Extraction -- 4.1 Local Color RGB Histograms -- 4.2 Local Color Invariance Histograms -- 5 Maximal Similarity Based Region Classification -- 6 Results and Discussion -- 6.1 Choice of the Appropriate Couple of CI/ET -- 6.2 MSRC and Descriptors Performance -- 6.3 Comparative Evaluation and Discussion -- 7 Conclusion -- References -- Joint Segmentation of Myocardium on Rest and Stress Spect Images -- 1 Introduction -- 2 Segmentation -- 2.1 Three Phase Local Binary Fitting Model -- 2.2 Parallelism and Distance Constraint -- 2.3 Shape Constraint -- 2.4 Joint Segmentation -- 3 Results and Conclusion -- 3.1 Qualitative Evaluation -- 3.2 Quantitative Evaluation -- 3.3 Conclusion -- References -- Parallel Hough Space Image Generation Method for Real-Time Lane Detection -- Abstract -- 1 Introduction -- 2 Lane Detection System -- 2.1 Lane Feature Point Extraction -- 2.2 Standard Hough Transform -- 2.3 Compact Hough Transform -- 3 Experimental Results -- 4 Conclusions -- Acknowledgment -- References -- A Novel Decentralised System Architecture for Multi-camera Target Tracking -- 1 Introduction -- 2 Decentralised Multi-camera Tracking System -- 2.1 DA -- Target Detection -- 2.2 SS -- Target Information Storage -- 2.3 TA -- Single View Target Tracking -- 2.4 DS -- Data Collation and Multi-view Tracking -- 3 Multi-view IMS-SWAD Tracking Algorithm -- 3.1 Non-overlapping Camera Views -- 3.2 Overlapping Camera Views -- 4 Experimental Results
• Coral Reef Fish Detection and Recognition in Underwater Videos by Supervised Machine Learning: Comparison Between Deep Learning and HOG+SVM Methods -- 1 Introduction -- 2 Presentation of the Methods -- 2.1 Histogram of Oriented Gradients + Support Vector Machine -- 2.2 Deep Learning -- 3 Practical Implementation -- 3.1 Data Preprocessing -- 3.2 Detection/Recognition Pipeline -- 3.3 Post-processing and Bounding Box Fusion -- 4 Results, Comparison and Discussion -- 5 Future Work -- References -- A Real-Time Eye Gesture Recognition System Based on Fuzzy Inference System for Mobile Devices Monitoring -- 1 Introduction -- 2 Related Work -- 3 Proposed System -- 3.1 Face and Pupils Detection -- 3.2 Eye Gesture Recognition -- 4 Experiment Result -- 5 Conclusion -- References -- Spatially Varying Weighting Function-Based Global and Local Statistical Active Contours. Application to X-Ray Images -- 1 Introduction -- 2 Background -- 2.1 Active Contours -- 2.2 External Forces -- 3 The Proposed Model -- 3.1 Progressing the Model Curve -- 4 Experiments -- 5 Conclusion -- References -- Vegetation Segmentation in Cornfield Images Using Bag of Words -- 1 Introduction -- 2 Literature Review -- 3 Proposed Methodology -- 3.1 Classification Model -- 3.2 Image Vegetation Segmentation -- 4 Experimental Results -- 4.1 Comparative Analysis -- 5 Conclusions -- References -- Fast Traffic Sign Recognition Using Color Segmentation and Deep Convolutional Networks -- 1 Introduction -- 2 Related Work -- 3 Proposed Approach -- 4 The Data Set of Italian Traffic Signs -- 5 Experimental Results -- 6 Conclusions -- References -- The Orlando Project: A 28nm FD-SOI Low Memory Embedded Neural Network ASIC -- 1 Introduction -- 2 Overview of the State of the Art -- 3 Orlando System-On-a-Chip -- 4 Model Compression -- 4.1 Compression Strategy -- 5 Experiments -- 5.1 Classification
• 3 Proposed Approach
• 4.1 Tracking Performance -- 4.2 Best View Selection and Occlusion Handling -- 5 Conclusion -- References -- Intramolecular FRET Efficiency Measures for Time-Lapse Fluorescence Microscopy Images -- 1 Introduction -- 2 Quantitative FRET Imaging Model -- 3 Measuring FRET Efficiency -- 4 Uncertainty of FRET Efficiency Measure -- 5 Evaluation -- 6 Conclusion -- References -- Predicting Image Aesthetics with Deep Learning -- 1 Introduction -- 2 Database and Evaluation Criterions -- 3 Proposed Approach for Image Aesthetic Assessment -- 3.1 DeepIA: A CNN for Image Aesthetic Assessment -- 4 Experimental Results -- 5 Conclusions -- References -- Automatic Image Splicing Detection Based on Noise Density Analysis in Raw Images -- 1 Introduction -- 1.1 Image Splicing Detection -- 1.2 Splicing Detection from Image Noise -- 2 Noise Density Contribution Tables -- 2.1 Principles and Definitions -- 2.2 Noise Density Contribution Table -- 2.3 Classification Between Type and Type -- 3 Seed Expansion -- 4 Multichannel -- 5 Implementation and Results -- 6 Conclusion -- References -- Breast Shape Parametrization Through Planar Projections -- 1 Introduction -- 2 Thin-Plate Spline in Biomedical Investigations -- 3 Materials and Methods -- 3.1 Dataset Acquisition -- 3.2 Standard Projection of the Breast Shape -- 4 Results -- 5 Conclusions -- References -- Decreasing Time Consumption of Microscopy Image Segmentation Through Parallel Processing on the GPU -- 1 Introduction -- 2 CPU-GPU Parallel Processing -- 3 Microscopy Image Segmentation -- 3.1 Notations -- 3.2 Superpixel Segmentation -- 3.3 Active Contour Segmentation -- 4 Accelerating Programs Using the GPU -- 4.1 Strong and Weak Scaling -- 4.2 Estimating the Achievable Speedup -- 5 Conclusion -- References