Pattern Recognition Applications and Methods: 12th International Conference, ICPRAM 2023, Lisbon, Portugal, February 22–24, 2023, Revised Selected Papers (Lecture Notes in Computer Science)

✍ Scribed by Maria De Marsico (editor), Gabriella Sanniti Di Baja (editor), Ana Fred (editor)

• Publisher: Springer
• Year: 2024
• Tongue: English
• Leaves: 159
• Category: Library

✦ Synopsis

This book constitutes the thoroughly refereed and revised selected papers from the 12th International Conference on Pattern Recognition Applications and Methods, ICPRAM 2023, held in Lisbon, Portugal, during February 22–24, 2023.

The 28 full papers and 42 short papers included in this book were carefully reviewed and selected from 157 submissions. They were organized in topical sections as follows: theory and methods and applications.

✦ Table of Contents

Preface
Organization
Contents
Theory and Methods
Exploring Data Augmentation Strategies for Diagonal Earlobe Crease Detection
1 Introduction
2 Related Work
2.1 Physiological Perspectives on DELC
2.2 Computer Vision Perspectives on DELC
3 Dataset
4 Proposal Description
4.1 Proposed Architecture
4.2 Data Augmentation
4.3 Backbone Comparison
5 Experimental Setup
6 Results
7 Conclusions
References
Distance Transform in Images and Connected Plane Graphs
1 Introduction
2 Background and Definitions
3 DT in a Connected Plane Graph
3.1 Bottom-Up Construction in the Irregular Pyramid
3.2 Top-Down Propagation
4 Distance Transform (DT) in a Binary Image
5 Evaluation and Results
6 Conclusion
References
Real-World Indoor Location Assessment with Unmodified RFID Antennas
1 Introduction
2 State of the Art
3 RFID Antenna Description
4 Previous Solution
5 Scenario
6 Evolved Solution
6.1 Baseline
6.2 First Pre-processing: IQR
6.3 Second Pre-processing: AutoEncoder
7 Conclusion
References
Applications
MSAA-Net: Multi-Scale Attention Assembler Network Based on Multiple Instance Learning for Pathological Image Analysis
1 Introduction
2 Related Work
2.1 WSI Classification
2.2 MIL-Based WSI Classification
3 Proposed Method
3.1 Problem Formulation
3.2 Multi-scale Attention Assembler Network
4 Experiment
4.1 Dataset
4.2 Comparison Methods and Evaluation Metrices
4.3 Implementation Details
4.4 Results
5 Discussion
5.1 Visualization of Attention Maps
5.2 Distribution of Attention Weights of Each Method
5.3 Attention Weight Assignment for Each Scale of MSAA-Net
6 Conclusion
References
Analysis of Generative Data Augmentation for Face Antispoofing
1 Introduction
2 Similar Work
3 Proposed Method
3.1 Preprocessing
3.2 Data Augmentation
3.3 Model Training
4 Experiment Setup
4.1 Datasets
4.2 Dataset Augmentation
4.3 Face Antispoofing
4.4 Keyframe Selection
5 Results
6 Conclusion
References
Improving Person Re-identification Through Low-Light Image Enhancement
1 Introduction
2 Related Work
3 ReID Pipeline
4 Experiments and Results
4.1 ReID on the Original Dataset
4.2 ReID on Footage Improved with Retinex Algorithms
4.3 ReID on Footage Improved with Deep Learning Approaches
4.4 Ablation Study: Position of the Enhancement Stage
5 Discussion
6 Conclusions
References
Gender-Aware Speech Emotion Recognition in Multiple Languages
1 Introduction
2 Constructing the Multilingual SER Dataset
3 The Considered Classification Models
3.1 k-NN
3.2 Transfer Learning Based on YAMNet
3.3 Bidirectional LSTM
4 Experimental Protocol and Results
4.1 Performance Analysis of Each Classifier Across Different Languages
5 Conclusion and Future Developments
References
Pattern Recognition Techniques in Image-Based Material Classification of Ancient Manuscripts
1 Introduction
1.1 Dead Sea Scrolls Collection
1.2 Pattern Recognition Techniques
2 Methodology
2.1 Classification Using Fourier Transform
2.2 Hierarchical K-Means Clustering
2.3 Convolutional Neural Networks
3 Results
3.1 Classification Using Fourier Transform
3.2 Hierarchical K-Means Clustering
3.3 Convolutional Neural Networks
4 Discussions
5 Conclusions
References
Author Index