Image and Video Technology

Name: Image and Video Technology | PSIVT 2017 International Workshops, Wuhan, China, November 20-24, 2017, Revised Selected Papers
Brand: Springer
Price: 53.49 EUR
Availability: OnlineOnly

PSIVT 2017 International Workshops, Wuhan, China, November 20-24, 2017, Revised Selected Papers

Shin'ichi Satoh(Herausgeber*in)

Springer (Verlag)

Erschienen am 7. Juni 2018

X, 480 Seiten

E-Book

PDF mit Wasserzeichen-DRM

Systemvoraussetzungen

978-3-319-92753-4 (ISBN)

53,49 €inkl. 7% MwSt.

Systemvoraussetzungen

für PDF mit Wasserzeichen-DRM

E-Book Einzellizenz

Als Download verfügbar

Beschreibung

Weitere Details

Weitere Ausgaben

Inhalt

Intro
Preface
Contents
Human Behaviour Analysis
Workshop on Human Behavior Analysis
Organization
Deep Transfer Feature Based Convolutional Neural Forests for Head Pose Estimation
1 Introduction
2 Deep Transfer Feature Based Convolution Neural Forests for Head Pose Estimation
2.1 Deep Transfer Feature Representation
2.2 D-CNF Training
2.3 D-CNF for Head Pose Estimation
3 Experimental Results
3.1 Datasets and Settings
3.2 Experiments on Pointing'04 Datasets
3.3 Experiments on Multi-view BU3D-HP Dataset
3.4 Experiments on CCNU-HP Dataset in the Wide Scene
4 Conclusion
References
Biometric System Based on Registration of Dorsal Hand Vein Configurations
1 Introduction
2 The Database
3 The Proposed System
3.1 Inhomogeneity Correction
3.2 Preprocessing
3.3 Vein Segmentation
3.4 Vein Registration
3.5 Similarity Evaluation
3.6 Classification
4 Results and Experiments
5 Conclusion and Future Work
References
A Multi-scale Triplet Deep Convolutional Neural Network for Person Re-identification
1 Introduction
2 The Proposed MST-CNN Architecture
2.1 Formulation
2.2 The Network Architecture
2.3 Triplet Selection
2.4 L2 Normalization
3 Experimental Results
3.1 The Datasets
3.2 The Evaluation Protocol
3.3 Training Strategy
3.4 Experimental Results on CUHK01 Dataset
3.5 Experimental Results on PRID450s Dataset
3.6 Single Scale Performance for Each Sub-CNN
4 Conclusion and in Future Work
References
Facial Expression Recognition Using Cascaded Random Forest Based on Local Features
1 Introduction
2 Cascaded RF Based on Local Features
2.1 Feature Extraction Based on Local Salient Region
2.2 Random Forest Algorithm
2.3 Cascaded Random Forest
3 Experiments
3.1 Settings
3.2 Comparison with State of the Art
3.3 Cross-Dataset Performance
4 Conclusions
References
Detection of Salient Regions in Crowded Scenes Based on Weighted Networks Approach
Abstract
1 Introduction
2 Crowd Weighted Network
2.1 Calculating Crowd Velocity Field
2.2 Representation of Crowd Network
3 Feature Extraction of Crowd Weighted Network
3.1 Calculating Node Strength
3.2 Normalization of Node Strength
3.3 Smooth Node Strength Field by Mean Filter Template
4 Experimental Results
4.1 Counter Flow Detection
4.2 Counter Flow Detection
4.3 Crowd Instability Detection
5 Conclusion
Acknowledgments
References
Selecting Salient Features from Facial Components for Face Recognition
1 Introduction
2 Methods and Techniques
2.1 Preprocessing
2.2 Component Detection
2.3 Feature Extraction
2.4 Feature Selection
2.5 Feature Normalization
3 Experiment Analysis
3.1 Facial Databases
3.2 Results and Performance Evaluation
3.3 Comparison with the State of the Art Methods
4 Conclusion and Future Work
References
Educational Cloud and Image and Video Enriched Cloud Services
Workshop on Educational Cloud and Imageand Video Enriched Cloud Services
Organization
Research on Optimization of Point Cloud Registration ICP Algorithm
Abstract
1 Introduction
2 ICP Algorithm and the Improved Method
2.1 Classic ICP Algorithm
2.2 Analysis of ICP Algorithm
2.3 Improved ICP Algorithm
3 Improved ICP Algorithm Test
4 Conclusion
Acknowledgments
References
High School Statistical Graph Classification Using Hierarchical Model for Intelligent Mathematics Problem Solving
1 Introduction
2 The Proposed Method
3 Experimental Results and Analysis
3.1 Dataset Description
3.2 Experimental Results
4 Conclusion
References
Machine Solving on Hypergeometric Distribution Problems
1 Introduction
2 Related Work
3 Hypergeometric Distribution Problems
4 Our Method
4.1 Problem Type Determination
4.2 Data Extraction and Processing
4.3 Solution to Hypergeometric Distribution Problem
5 Experiments
5.1 Experimental Data
5.2 Experimental Results and Analysis
5.3 An Illustration
6 Conclusion
References
Extracting Algebraic Relations from Circuit Images Using Topology Breaking Down and Shrinking
Abstract
1 Introduction and Related Work
2 Algorithm Based on Topology Shrinking
3 Circuit Graph Generation
3.1 Components Recognition
3.2 Graph Node Analysis
4 Algebraic Relations Extraction
4.1 Topology Breaking Down
4.2 Topology Shrinking
5 Experimental Results
5.1 Dataset Preparation and Experiment Settings
5.2 Experiments on Relations Extraction
6 Conclusions and Future Work
Acknowledgment
References
Parallel Education Systems Under Perspective of System Construction for New IT Era
Abstract
1 New IT Era and Challenges in Education
2 Development of Constructivism
2.1 History
2.2 Constructivism from Perspective of Cybernetics
3 Personality Construction
4 System Construction Based on Knowledge Construction and Psychological Construction
5 ACP and Parallel Education System Framework from Perspective of System Construction
5.1 ACP Based on Parallel Systems
5.2 Parallel Education System Framework from Perspective of System Construction
5.3 Key Technologies
6 Conclusions
Acknowledgments
References
Foot Modeling Based on Machine Vision and Social Manufacturing Research
Abstract
1 Introduction
2 Social Manufacturing of the Footwear Industry
2.1 Introduction to Social Manufacturing
2.2 Social Manufacturing System of Footwear Industry
3 Machine Vision Technology Used in Footwear Society Manufacturing
3.1 The Use of Binocular Vision for the Establishment of Three-Dimensional Model
3.2 Camera Calibration
3.3 Experimental Results
4 Conclusion
Acknowledgments
References
Computerized Adaptive English Ability Assessment Based on Deep Learning
1 Introduction
2 Different CAT Models
2.1 Conventional CAT Model Based on IRT
2.2 CAT Based on Deep Learning
3 Settings of Deep Neural Networks for CAT
3.1 Performance Index
3.2 Decision Variables
3.3 Training Algorithms of Neural Networks
4 Testing Results
4.1 Testing Data
4.2 Improved IRT Testing Results
4.3 The Appropriate Structure of Model-GRU
4.4 Comparison Between Conventional Improved CAT and Model-GRU
5 Conclusion
References
China-Finland EduCloud Platform Towards Innovative Education
Abstract
1 Introduction
2 Innovative Education
3 Innovative Education Ecosystem
4 Challenges for Innovative Education
5 EduCloud: Intelligent Cloud-Based Education Platform
5.1 Cloud Computing Overview
5.2 EduCloud Computing
6 EduCloud-Enhanced Education
7 EduCloud Practices and Cooperation Initiatives
7.1 CCNU EduCloud Practices
7.2 Finpeda Virtual Space
7.3 China-Finland EduCloud Cooperation Initiatives
8 Conclusion and Future Work
References
SPSE: A Smart Phone-Based Student Evaluation
Abstract
1 Introduction
2 Related Works
3 Smart Phone-Based Student Evaluation (SPSE)
3.1 SPSE Model
3.2 Dynamic Learning Behavior Analysis
3.3 Static Student Scoring
3.4 Comprehensive Student Evaluation
4 Conclusion
Acknowledgement
References
Research on the Construction of Corpus for Automatic Solution of Elementary Mathematics Statistics Applications
Abstract
1 Introduction
2 Related Research
2.1 The Construction of Corpus
2.2 The Automatic Solution of Mathematics Applications
2.3 The Construction of Applications Corpus
3 The Construction of Corpus for Automatic Solution of Elementary Mathematics Statistics Applications
3.1 The Source of the Corpus
3.2 The Classification System of Elementary Mathematics Statistics
3.3 The Metadata of Corpus Entry
3.4 Semantic Annotation and Representation of Corpus
4 The Implementation of Corpus
4.1 The Implementation of Corpus Indexing Tool
4.2 Statistical Information of the Corpus
5 Conclusion
Acknowledgements
References
Constructing a Learning Map with Lattice for a Programming Course
1 Introduction
2 Related Work
2.1 Algorithms for generating lattices
2.2 Online Learning System
2.3 Visual Analytics
3 Preliminaries
4 Constructing Knowledge Map
4.1 Attribute Extraction
4.2 Concept Construction
4.3 Knowledge Map Construction
5 Knowledge Map Application
5.1 Attribute Analysis
5.2 Attribute Relationship
5.3 Knowledge Map
6 Conclusion
References
Data Mining as a Cloud Service for Learning Artificial Intelligence
1 Introduction
2 DMCS Architecture
3 DMCS Implementation
3.1 Big Data Framework for DMCS
3.2 DMCS Library for Machine Learning
3.3 DMCS User-Friendly Interface
4 DMCS Demonstration
5 Discussion
6 Conclusions and Future Work
References
Vision Meets Graphics
Workshop on Vision Meets Graphics(VG 2017)
Organization
Crowd Counting from a Still Image Using Multi-scale Fully Convolutional Network with Adaptive Human-Shaped Kernel
1 Introduction
2 Related Work
3 Proposed Method
3.1 Generating the Ground Truth of Density Map
3.2 Multi-scale Fully Convolutional Network for Crowd Counting
3.3 Details
4 Experimental Analysis
4.1 Evaluation Metrics
4.2 Evaluations of MSFCN
4.3 Comparisons with State-of-the-art Approaches
5 Conclusion
References
On Road Vehicle Detection Using an Improved Faster RCNN Framework with Small-Size Region Up-Scaling Strategy
1 Introduction
2 Proposed Method
2.1 Introduction of Traditional Faster RCNN
2.2 Improvements of Traditional Faster RCNN
2.3 Small Region Up-Sampling Strategy
3 Experimental Analysis
3.1 Implementation Details and Evaluation Indicators
3.2 Evaluations in VOC2007 Dataset
3.3 Evaluations in Kitti Dataset
3.4 Evaluations in Our Own Dataset
4 Conclusions and Future Work
References
Fast Haze Removal of UAV Images Based on Dark Channel Prior
1 Introduction
2 Background
2.1 Atmospheric Scattering Model
2.2 Dark Channel Prior
2.3 Dehazing Algorithm Using Dark Channel Prior
3 Fast Haze Removal Algorithm Based on Dark Channel Prior
3.1 Estimates Atmospheric Light Using the Quad-Tree Subdivision
3.2 Improve Transmission Acquisition
4 Experimental Analysis and Comparison
4.1 Subjective Analyses
4.2 Objective Evaluation
5 Conclusion
References
Watercolour Rendering of Portraits
1 Introduction
2 Related Work
3 Our Approach
3.1 Image Brightening
3.2 Wobbling
3.3 Glazing
3.4 Pigment Diffusion
3.5 Level of Detail
4 Experiments
5 Conclusions
References
Blind Image Deblurring via Salient Structure Detection and Sparse Representation
1 Introduction
1.1 Related Work
1.2 Our Contributions
2 Kernel Estimation
2.1 A Brief Introduction to Sparse Representation
2.2 Interim Latent Image Estimation via Sparse Model
2.3 Image Structure Computation
2.4 Kernel Estimation via Sparse Model
3 Latent Image Estimation by 1-Norm Deconvolution
4 Initializing Our Algorithm
5 Experiments
5.1 Dictionary Learning
5.2 Experimental Results and Evaluation
5.3 Evaluation on the Dataset of LevinspsCVPR2009
6 Conclusions
References
Blur Estimation for Natural Edge Appearance in Computational Photography
1 Introduction
2 Basics and Notations
2.1 Just Noticeable Defocus Blur Estimation
2.2 Image Segmentation and Meanshift Filter
2.3 Stereo Vision
3 Method
3.1 Meanshift Filter
3.2 Simple Light Control Based on Stereo Vision
4 Experiments and Discussion
4.1 Meanshift Filter
4.2 Depth-Related Darkening
4.3 Defocus Blur
5 Conclusions
References
Structure-Preserving Texture Smoothing via Adaptive Patches
1 Introduction
2 Related Work
2.1 Edge-Preserving Image Smoothing
2.2 Structure-Preserving Texture Smoothing
2.3 SLIC Superpixel Segmentation
3 The Proposed Method
3.1 Guidance Image Computation via Superpixels
3.2 Guided Bilateral Filtering
4 Analysis
5 Results and Comparisons
6 Applications
7 Conclusions and Future Works
References
Robust Blind Deconvolution Using Relative Total Variation as a Regularization Penalty
Abstract
1 Introduction
2 Related Work
3 Robust Blind Kernel Estimation
3.1 Intermediate Latent Image Estimation
3.2 Salient Edge Selection
3.3 Kernel Estimation
4 Image Recovery
5 Experimental Details and Results
6 Conclusion
Acknowledgement
References
Passive and Active Electro-Optical Sensors for Areal and Space Imaging
Passive and Active Electro-Optical Sensorsfor Aerial and Space Imaging (EO4AS)
Organization
An Aircraft Tracking Method in Simulated Infrared Image Sequences Based on Regional Distribution
Abstract
1 Introduction
2 Distribution Field
3 The Aircraft Tracking Algorithm
3.1 Region Proposal for Searching
3.2 Structural Feature Extraction
3.3 Model Matching and Update
3.4 Our Aircraft Tracking Algorithm
4 Experiments
4.1 Analyzing Regional Distribution Tracker
4.2 Evaluating Tracking Benchmark
5 Conclusion
Acknowledgments
References
DESIS - DLR Earth Sensing Imaging Spectrometer
1 Introduction
2 MUSES
3 DESIS Design
3.1 Spectrometer Optic
3.2 Focal Plane Assembly
3.3 Instrument Control and Power Supply
3.4 Calibration Unit
3.5 Pointing Unit
3.6 Container and Thermal Design
4 First Calibration Results
4.1 MTF
4.2 Linearity
4.3 Photo Transfer Curve and Conversion Gain
5 Conclusion and Outlook
References
FireBIRD Mission Data for Gas Flaring Analysis
Abstract
1 Introduction
2 Remote Sensing of Gas Flaring
3 FireBIRD Mission
3.1 Sensor Specifications
3.2 Fire Products
3.3 Fire Processor Validation for the Case of Gas Flares
4 Summary and Outlook
References
Automatic Ship Detection on Multispectral and Thermal Infrared Aerial Images Using MACS-Mar Remote Sensing Platform
Abstract
1 Introduction
2 Methodology
2.1 Automatic Object Detection
2.2 Accuracy Assessment
3 Experiment
3.1 MACS - Modular Aerial Camera System
3.2 Joint EMSec Experiment
3.3 Thermal Imagery Experiment
4 Results
5 Discussion
6 Future Work
Acknowledgments
References
Extracting Plücker Line and Their Relations for 3D Reconstruction of Indoor Scene
1 Introduction
2 Related Work
3 Overview of the Proposed Method
4 The Related Properties of Plücker Line
4.1 Introduction of Plücker Coordinate
4.2 The Presentation of Spatial Screw Motion of Robot
5 Experiments and Results
6 Conclusion and Future Work
References
Computer Vision and Modern Vehicles
Workshop on Computer Vision and ModernVehicles (CVMV 2017)
Organization
Context-Awareness Based Adaptive Gaussian Mixture Background Modeling
Abstract
1 Introduction
2 Gaussian Mixture Model
3 Context-Awareness Based Gaussian Mixture Models
4 Experimental Results and Discussion
5 Conclusions
Acknowledgment
References
Robust Expression Recognition Using ResNet with a Biologically-Plausible Activation Function
1 Introduction
2 Background
2.1 Response Function
2.2 Noisy Softplus (NSP)
3 Methods
3.1 Residual Learning
3.2 Network Architecture
4 Experiments
4.1 Datasets
4.2 Results and Analysis
4.3 Power Consumption Evaluation
5 Conclusions
References
Local Fast R-CNN Flow for Object-Centric Event Recognition in Complex Traffic Scenes
1 Introduction
2 Related Work
3 Methodology
3.1 Overview
3.2 Region-Based Convolutional Neural Network
3.3 Local Fast R-CNN Flow
4 Experiments
5 Conclusions
References
Mixed-Noise Removal in Images Based on a Convolutional Neural Network
1 Introduction
2 Image Super-Resolution Algorithm
2.1 Super-Resolution Convolutional Neural Network
2.2 FSRCNN
3 Improved CNN-Based Denoising Algorithm
4 Experimental Results and Analysis
4.1 Data and Testbed
4.2 Comparing with DDID, AMF and SSTV
4.3 Subjective Results
4.4 Objective Results
5 Conclusions
References
A Systematic Scheme for Automatic Airplane Detection from High-Resolution Remote Sensing Images
Abstract
1 Introduction
2 Airport Detection
2.1 Down-Sample and FT Saliency Detection
2.2 Airport Region Locating Based on LSD
3 Airplane Detection
3.1 Candidate Location Extraction by CF-Filter
3.2 Identify Airplanes by CNNs Model
4 Experiment and Analysis
5 Conclusion and Future Works
References
Author Index

Systemvoraussetzungen

Als PDF speichern Als Link merken