Chapter 9 Implementation of deep learning techniques on thermal image classification
-
Prosenjit Chatterjee
Abstract
Thermal imaging is vital across various domains, including security, healthcare, and industrial applications. However, achieving accurate classification of thermal images remains challenging due to the inherent complexity of thermal image data and the limited availability of annotated datasets. This chapter presents a practical approach leveraging deep learning architectures for effective thermal image classification. Pretrained convolutional neural networks (CNNs) such as AlexNet, ZFNet, GoogLeNet, VGGNet-19, and ResNet-50 were fine-tuned using transfer learning on annotated datasets. The findings reveal that deeper CNN architectures, notably VGGNet- 19 and ResNet-50, achieved superior classification accuracy, with robust performance observed on the Tufts Thermal Face dataset. Integrating a Kalman filter as a preprocessing step significantly reduced thermal image noise, enhancing model precision and effectiveness. These results underscore the potential of deep learning methodologies for thermal image analysis, especially in scenarios requiring reliable and accurate categorization. The study highlights the need for further research to adapt and extend this methodology to diverse datasets and real-world environments, enabling broader applications and advancing the capabilities of thermal imaging technology.
Abstract
Thermal imaging is vital across various domains, including security, healthcare, and industrial applications. However, achieving accurate classification of thermal images remains challenging due to the inherent complexity of thermal image data and the limited availability of annotated datasets. This chapter presents a practical approach leveraging deep learning architectures for effective thermal image classification. Pretrained convolutional neural networks (CNNs) such as AlexNet, ZFNet, GoogLeNet, VGGNet-19, and ResNet-50 were fine-tuned using transfer learning on annotated datasets. The findings reveal that deeper CNN architectures, notably VGGNet- 19 and ResNet-50, achieved superior classification accuracy, with robust performance observed on the Tufts Thermal Face dataset. Integrating a Kalman filter as a preprocessing step significantly reduced thermal image noise, enhancing model precision and effectiveness. These results underscore the potential of deep learning methodologies for thermal image analysis, especially in scenarios requiring reliable and accurate categorization. The study highlights the need for further research to adapt and extend this methodology to diverse datasets and real-world environments, enabling broader applications and advancing the capabilities of thermal imaging technology.
Kapitel in diesem Buch
- Frontmatter I
- Preface V
- Contents VII
-
Section: Image processing
- Chapter 1 Magnetic resonance image re-parameterization on real data 1
- Chapter 2 Denoising and gradient fusion for effective edge detection for noisy color images 17
- Chapter 3 Understanding driver attention to objects for ADASs: what do drivers see? 39
- Chapter 4 Image clustering enhanced with refined image classification 59
- Chapter 5 AI-powered framework for objective scoring of product design innovation 89
-
Section: Computer vision
- Chapter 6 Image inpainting using GAN transformerbased model 111
- Chapter 7 Enhanced image watermarking through cross-attention and noise-invariant domain learning 127
- Chapter 8 Online melt pool monitoring using a deep transformer image processing solution 153
- Chapter 9 Implementation of deep learning techniques on thermal image classification 173
- Chapter 10 Drishti: a generative AI-based application for gesture recognition and execution 203
-
Section: Pattern recognition
- Chapter 11 Exploring muzzle biometrics: a deep learning framework for noninvasive cattle recognition 239
- Chapter 12 Utilizing real-world data to develop a userindependent sensor-based human activity recognition system 253
- Index 273
Kapitel in diesem Buch
- Frontmatter I
- Preface V
- Contents VII
-
Section: Image processing
- Chapter 1 Magnetic resonance image re-parameterization on real data 1
- Chapter 2 Denoising and gradient fusion for effective edge detection for noisy color images 17
- Chapter 3 Understanding driver attention to objects for ADASs: what do drivers see? 39
- Chapter 4 Image clustering enhanced with refined image classification 59
- Chapter 5 AI-powered framework for objective scoring of product design innovation 89
-
Section: Computer vision
- Chapter 6 Image inpainting using GAN transformerbased model 111
- Chapter 7 Enhanced image watermarking through cross-attention and noise-invariant domain learning 127
- Chapter 8 Online melt pool monitoring using a deep transformer image processing solution 153
- Chapter 9 Implementation of deep learning techniques on thermal image classification 173
- Chapter 10 Drishti: a generative AI-based application for gesture recognition and execution 203
-
Section: Pattern recognition
- Chapter 11 Exploring muzzle biometrics: a deep learning framework for noninvasive cattle recognition 239
- Chapter 12 Utilizing real-world data to develop a userindependent sensor-based human activity recognition system 253
- Index 273
