Third International Conference on Image Processing and Capsule Networks : ICIPCN 2022 / Joy Iong-Zong Chen, João Manuel R. S. Tavares, Fuqian Shi, editors.

This book provides a collection of the state-of-the-art research attempts to tackle the challenges in image and signal processing from various novel and potential research perspectives. The book investigates feature extraction techniques, image enhancement methods, reconstruction models, object dete...

Full description

Saved in:
Bibliographic Details
Corporate Author: International Conference on Image Processing and Capsule Networks Online
Other Authors: Chen, Joy Iong-Zong (Editor), Tavares, João Manuel R. S. (Editor), Shi, Fuqian (Editor)
Format: eBook
Language:English
Published: Cham, Switzerland : Springer, 2022.
Series:Lecture notes in networks and systems ; v. 514.
Subjects:
Online Access:Click for online access
Table of Contents:
  • Intro
  • Preface
  • Contents
  • Brain-Inspired Spatiotemporal Feature Extraction Using Convolutional Legendre Memory Unit
  • 1 Introduction
  • 1.1 Neuromorphic Computing
  • 2 Related Works
  • 3 Proposed Convolutional LMU Model
  • 4 Synthetic Dataset and Evaluation Measures
  • 5 Results and Analysis
  • 6 Conclusion
  • References
  • Underwater Image Enhancement Using Image Processing
  • 1 Introduction
  • 1.1 Problem Statement
  • 2 Literature Survey
  • 3 Methodology
  • 4 Architecture
  • 5 Conclusion
  • References
  • Fake News Detection on Indian Sources
  • 1 Introduction
  • 2 Related Works
  • 3 Proposed Solution
  • 3.1 Dataset
  • 3.2 Data Cleaning
  • 3.3 Data Analysis
  • 3.4 Text Preprocessing and Text Transformation
  • 3.5 Model
  • 3.6 Testing
  • 4 Results
  • 5 Use Cases
  • 6 Future Works
  • 7 Conclusion
  • References
  • Exploring Self-supervised Capsule Networks for Improved Classification with Data Scarcity
  • 1 Introduction
  • 2 Related Work
  • 2.1 Functionality of Capsule Networks
  • 2.2 Self-supervision and Capsule Networks
  • 2.3 Pretrained Capsule Networks
  • 3 Methods
  • 3.1 Data Set
  • 3.2 Capsule Network Model
  • 3.3 Self-supervision
  • 4 Results and Discussion
  • 4.1 Data Scarcity
  • 4.2 Learning Behaviour of the Self-supervised CapsNet
  • 4.3 Data Scarcity and Imbalance
  • 4.4 Correlation of Pretext and Downstream Accuracy
  • 5 Conclusion
  • References
  • A Novel Architecture for Improving Tuberculosis Detection from Microscopic Sputum Smear Images
  • 1 Introduction
  • 2 Literature Review
  • 3 Methodology
  • 3.1 Preprocessing
  • 3.2 Mask Generation Using SegZNet Architecture
  • 3.3 Data Augmentation
  • 3.4 UNet Segmentation
  • 4 Result and Discussion
  • 5 Conclusion
  • References
  • TapasQA
  • Question Answering on Statistical Plots Using Google TAPAS
  • 1 Purpose
  • 2 Previous Work
  • 3 Methodology
  • 3.1 Dataset
  • 3.2 Pipeline
  • 4 Major Research Findings
  • 4.1 Questions Handled by Our Model
  • 4.2 Training Details
  • 4.3 Evaluation Metric
  • 5 Result Implications
  • 5.1 Plot Element Detection Stage
  • 5.2 Table Question Answering (QA) Stage
  • 6 End-To-End Example
  • 7 Value and Limitations
  • 8 Conclusion and Future Work
  • References
  • Face Sketch-Photo Synthesis and Recognition
  • 1 Introduction
  • 2 Related Work
  • 3 Research Gap
  • 4 Data
  • 4.1 CUFS Dataset
  • 4.2 CelebA Dataset
  • 4.3 ORL Dataset
  • 5 Tools and Experimental Settings
  • 6 Proposed Methodology
  • 6.1 Face-Sketch Synthesis
  • 6.2 Face-Photo Synthesis
  • 6.3 Facial Recognition
  • 7 Results
  • 7.1 Face-Sketch Synthesis
  • 7.2 Face-Photo Synthesis
  • 7.3 Facial Recognition
  • 8 Evaluation
  • 8.1 Face-Sketch Synthesis
  • 8.2 Face-Photo Synthesis
  • 8.3 Facial Recognition
  • 9 Conclusion
  • 10 Future Work
  • References
  • Toward Robust Image Pre-processing Steps for Vehicle Plate Recognition
  • 1 Introduction
  • 2 The Proposed Deskew Approach
  • 3 Performance Evaluation
  • 4 Conclusion
  • References