Intelligent Multimedia Processing and Computer Vision: Techniques and Applications (Computing and Networks)

✍ Scribed by Shyam Singh Rajput, Chen Chen (editor), Karm Veer Arya (editor)

• Publisher: IET
• Year: 2023
• Tongue: English
• Leaves: 369
• Category: Library

✦ Synopsis

Intelligent multimedia involves the computer processing and understanding of perceptual input from speech, text, videos and images. Reacting to these inputs is complex and involves research from engineering, computer science and cognitive science. Intelligent multimedia processing deals with the analysis of images and videos to extract useful information for numerous applications including medical imaging, robotics, remote sensing, autonomous driving, AR/VR, law enforcement, biometrics, multimedia enhancement and reconstruction, agriculture, and security. Intelligent multimedia processing and computer vision have seen an upsurge over the last few years. With the increasing use of intelligent multimedia processing techniques in various sectors, the requirement for fast and reliable techniques to analyse and process multimedia content is increasing day by day.

Intelligent Multimedia Processing and Computer Vision: Techniques and applications reviews cutting edge research in the areas of intelligent multimedia processing and computer vision techniques and applications with a particular emphasis on interdisciplinary approaches and novel solutions. The book is aimed at practicing engineers, scientists, technology professionals, researchers and advanced students in the fields of multimedia processing and security, image processing, computer vision, biometrics, intelligent and smart technologies, machine learning and deep learning, and autonomous systems.

✦ Table of Contents

Cover
Contents
About the editors
Foreword
1 Introduction
2 State-of-the-art analysis of deep learning techniques for image segmentation
Abstract
2.1 Introduction
2.2 Outline of deep learning models
2.2.1 Convolutional neural networks
2.2.2 Recurrent neural network
2.2.3 Encoder–decoder architecture
2.2.4 Generative adversarial networks
2.3 Types of semantic segmentation architectures
2.3.1 Convolutional neural network-based architecture
2.3.2 Encoder–decoder convolutional models
2.4 Performance assessment metrics
2.5 Performance analysis
2.6 Summary
References
3 Biometric-based computer vision for boundless possibilities: process, techniques, and challenges
Abstract
3.1 Introduction
3.2 Need for biometric-based CV
3.2.1 Benefits of biometric authentication
3.3 Restricts access
3.4 Records timeliness
3.5 Enhances security measures
3.6 Replaces passwords
3.7 Minimizes the human error
3.8 Eases installation
3.8.1 Offers reasonable costs
3.9 Biometric-based CV: process, techniques, and challenges
3.9.1 Biometric-based CV process
3.9.2 Biometric-based CV techniques
3.9.3 Different biometric traits
3.9.4 Challenges of biometric-based CV techniques
3.10 Application areas
3.11 Selection criteria of suitable biometrics
3.12 Future of biometric-based CV
3.13 Summary
References
4 Channel refinement of fingerprint pre-processing models
Abstract
4.1 Introduction
4.1.1 Research contributions
4.2 Related work
4.2.1 Fingerprint enhancement
4.2.2 Fingerprint ROI segmentation
4.2.3 Attention mechanisms
4.3 Proposed method
4.3.1 Channel refinement unit
4.3.2 Introducing channel refinement unit into a fingerprint pre-processing model
4.4 Experimental set-up
4.4.1 Databases
4.4.2 Assessment criteria
4.5 Results: fingerprint enhancement
4.5.1 Enhancement of latent fingerprints
4.5.2 Enhancement of rural Indian fingerprints
4.5.3 Contrasting CRU with squeeze and excitation block
4.5.4 Application of CRU to various deep models for fingerprint enhancement
4.6 Contrasting CR-GAN with cutting-edge fingerprint enhancement methods leveraging GAN framework
4.7 Impact of CRU on fingerprint enhancement
4.7.1 Ridge structure preservation
4.7.2 Ablation study
4.7.3 Successful scenarios
4.7.4 Challenging scenarios
4.7.5 Fingerprint ROI segmentation
4.8 Conclusion
References
5 A review of deep learning approaches for video-based crowd anomaly detection
Abstract
5.1 Introduction
5.2 Challenges and taxonomy of VCAD
5.2.1 Challenges of VCAD
5.2.2 Taxonomy of video-based CAD
5.3 Deep learning techniques for VCAD
5.3.1 CNN-based VCAD
5.3.2 Sequence-to-sequence models for VCAD
5.3.3 Generative models for VCAD
5.3.4 Hybrid models for VCAD
5.4 Review of the datasets
5.5 Future research scope
5.6 Conclusions
References
6 Natural language and mathematical reasoning
Abstract
6.1 Introduction
6.2 Language, communication, and time
6.3 Linear time and determinism as condition for language
6.4 Entropy and synergic dissipative structures
6.5 Chaos in psycholinguistics
6.6 Fractals and language
6.6.1 Fractal patterns and categories
6.6.2 Fractal patterns in language
6.7 Fuzzy considerations of language
6.8 Cognitive linguistics
6.8.1 Cognitive semantics
6.9 Language and grammar
6.9.1 Qualitative distinction of elements
6.9.2 Cognitive grammar
6.9.3 Dynamic approaches
6.9.4 Cognitive approaches to grammaticalization
6.9.5 Pragmatic-driven approaches
6.10 Applications of cognitive linguistics
6.11 Chaos and linguistics
6.11.1 Thermodynamics: diffusive and synergetic processes
6.11.2 The linguistic reasoning laws
6.12 Thermodynamic rules for linguistic reasoning
6.12.1 Working hypothesis derivations and rules
6.12.2 Rules and findings
6.13 Conclusions and final remarks
References
7 AI and machine learning in medical data processing
Abstract
7.1 Introduction
7.2 Literature review
7.3 What is seizure?
7.4 Role of EEG and speech signal in seizure detection
7.5 Machine learning techniques in seizure detection
7.5.1 Database used
7.5.2 Methodology
7.5.3 Shannon entropy
7.5.4 SVM classifier
7.6 Results
7.7 Deep learning techniques in seizure detection
7.7.1 Methodology
7.8 Results
7.9 State-of-the-art comparison of different techniques
7.10 Future scope
References
8 Progress of deep learning in digital pathology detection of chest radiographs
Abstract
8.1 Introduction
8.2 Overview of the CXR pathology
8.2.1 Lung nodule
8.2.2 Emphysema
8.2.3 Pneumonia
8.2.4 COVID-19
8.2.5 Pneumothorax
8.2.6 Pulmonary tuberculosis
8.3 Functional aspects of DL techniques
8.3.1 Classification
8.3.2 Enhancement
8.3.3 Segmentation
8.4 Application design considerations
8.4.1 Data preparation
8.4.2 Using transfer learning
8.4.3 Measurements
8.4.4 Performance improvement techniques
8.5 Limitations and future directions
8.5.1 Limitations
8.5.2 Future directions
8.6 Conclusions
References
9 Computer vision and modern machine learning techniques for autonomous driving
Abstract
9.1 Introduction and challenges of autonomous driving
9.1.1 Autonomous levels of driving
9.1.2 Modular strategy for autonomous driving systems
9.1.3 Challenges of computer vision in autonomous driving systems
9.2 Visual perception and object detection for autonomous driving
9.2.1 Methods for detection
9.3 Visual perception and object tracking for autonomous driving
9.3.1 Methods for visual tracking
9.4 Visual perception and segmentation for autonomous driving
9.4.1 Methods of segmentation
9.5 Visual perception and deep reinforcement learning for autonomous driving
9.5.1 Method
9.6 3D scene analysis, sensors, datasets, and CARLA simulator for autonomous driving
9.6.1 3D scene analysis
9.6.2 Sensors
9.6.3 Datasets
9.6.4 CARLA simulator
9.7 Localization and mapping for autonomous driving
9.7.1 Mapping
9.7.2 Localization
9.7.3 Simultaneous localization and mapping (SLAM)
References
10 Dehazing and vision enhancement: challenges and future scope
Abstract
10.1 Introduction
10.2 Organizing the survey
10.3 Classification and comparison of haze removal techniques
10.3.1 Depth map estimation-based dehazing
10.3.2 Filtering-based dehazing
10.3.3 Fusion-based dehazing
10.3.4 Enhancement-based dehazing
10.3.5 Meta-heuristic method-based dehazing
10.3.6 Transform-based dehazing
10.3.7 Variational-based dehazing
10.3.8 Learning-based dehazing
10.4 Experiments and results
10.4.1 Metrics used
10.5 Conclusions
References
11 Machine learning and revolution in agriculture: past, present and future
Abstract
11.1 Introduction
11.2 Challenges and present scenario in agricultural applications
11.3 Need for ML in agricultural applications
11.4 State-of-the-art ML techniques in agricultural applications
11.4.1 Crop and fertilizer prediction
11.4.2 Land area-specific prediction of suitable crops
11.4.3 Crop yield prediction and climate change impact assessment
11.4.4 Extraction of fertilizer dosages for precision agriculture
11.4.5 Predicting fertilizer usage in agriculture production (crop yield-based on fertilizer consumption)
11.4.6 Seed quality classification
11.4.7 Soil property prediction
11.5 Application areas
11.5.1 Crop and soil monitoring
11.5.2 Insect and plant disease detection
11.5.3 Livestock health monitoring
11.5.4 Intelligent spraying
11.5.5 Automatic weeding
11.5.6 Aerial survey and imaging
11.5.7 Produce grading and sorting
11.6 Future trends
11.7 Summary
References
12 AIand ML-based multimedia processing for surveillance
Abstract
12.1 Introduction
12.1.1 Need for AI and ML in surveillance
12.1.2 Impact of AI on surveillance
12.1.3 Impact of ML on surveillance
12.1.4 Earlier attempts to surveillance
12.2 Understanding AI and ML in surveillance
12.2.1 Smarter systems, better results
12.3 AIand ML-based models for multimedia
12.3.1 Architectural models and technologies
12.3.2 AIand ML-based learning models
12.4 Multimedia processing for video surveillance
12.5 Image-processing working framework for surveillance systems
12.6 AIand ML-based surveillance systems
12.6.1 Physical intrusion detection system
12.6.2 Stereo-based approach
12.6.3 Single camera approach
12.6.4 Virtual fence
12.6.5 Suspicious behavior identification system (SBIS)
12.7 Applications of AI and ML in surveillance
12.8 Challenges and limitations of AI and ML in multimedia processing
References
13 Action recognition techniques
13.1 Introduction
13.2 Challenges in action recognition in visual surveillance
13.3 Recent advancements
13.4 Applications of action recognition
13.4.1 Video surveillance
13.4.2 Industrial control [17–20]
13.4.3 Autonomous driving [26–29]
13.4.4 Intelligent transportation [39–42]
13.4.5 Human–computer interactions [52–55]
13.4.6 Visual appearance
13.4.7 Depth sensors [64–67]
13.4.8 Graph convolutional networks
13.5 Challenges and future directions
References
14 Conclusion
Index
Back Cover