Infrared Image Processing With Artificial Intelligence Progress And

Bonisiwe Shabane

-Dec 28, 2025, 3:29 AM

infrared image processing with artificial intelligence progress and

This special issue belongs to the section “AI in Imaging“. Infrared (IR) imaging has become an indispensable tool in a wide range of domains, including surveillance, remote sensing, medical diagnostics, autonomous driving, industrial inspection, and environmental monitoring. Unlike visible-spectrum imaging, infrared captures thermal radiation, offering unique advantages in low-light and obscured environments. However, IR image processing poses distinct challenges, such as low contrast, limited texture, noise, and variability in environmental conditions. Recent advances in artificial intelligence (AI)—particularly deep learning—have led to significant breakthroughs in addressing these challenges. AI-powered methods now play a crucial role in tasks such as image enhancement, object detection and tracking, semantic segmentation, anomaly detection, and multimodal fusion involving IR data.

Despite these advances, the field continues to face various obstacles, including limited labeled data, high domain variability, a lack of generalization, and the need for interpretable and energy-efficient models. This Special Issue aims to compile state-of-the-art research at the intersection of AI and infrared image processing. We welcome the submission of original contributions that explore novel algorithms, datasets, applications, and theoretical frameworks. Both supervised and unsupervised learning paradigms, multimodal approaches, and cross-domain adaptation methods are welcome. The scope of this Special Issue includes, but is not limited to, the following topics: A not-for-profit organization, IEEE is the world's largest technical professional organization dedicated to advancing technology for the benefit of humanity.© Copyright 2025 IEEE - All rights reserved.

Use of this web site signifies your agreement to the terms and conditions. The ability of generative models to accurately fit data distributions has resulted in their widespread adoption and success in fields such as computer vision and natural language processing. In this chapter, we provide a brief overview of the application of generative models in the domain of infrared (IR) image super-resolution, including a discussion of the various challenges and adversarial training methods employed. We propose potential areas for further investigation and advancement in the application of generative models for IR image super-resolution. In modern society, IR images play an irreplaceable role in industry and daily life. Although visible images make it easier to transfer information to people, in specific environments, such as earthquake rescue and security, where there is insufficient light, people have to turn to infrared images.

Compared to visible images, infrared images can be better tolerated in tough natural environments and provide rich information about heat sources. Such valuable feedback can help people judge the condition from outdoor equipment and individuals in order to repair damaged infrastructure or help someone in distress. Considering these important applications, high-resolution (HR) IR images are needed urgently. However, IR image resolution is unsatisfactory due to the limitations by current optical devices. With the growing interest in Generative Adversarial Networks (GANs)goodfellow2020generative in the deep learning community, introducing adversarial training methods for IR image super-resolution is on the agendaledig2017photo ; wang2018esrgan ; ma2020structure ; gulrajani2017improved . In this section, we will present IR imaging applications first, and some key domain will be discussed.

Then the basic components and challenges faced by infrared imaging systems are presented. IR image super-resolution is an attractive approach in a wide range of realistic situations. We will present some typical fields of them, such as medical engineering and engineering tasks. Then, other methods will be briefly discussed. IR image super-resolution applications can be seen in Fig.1 with more details. In the medical engineering field, degenerative diseases in the nervous system have been gaining great attention.

If the mechanisms behind such diseases can be clarified, we will have the opportunity to completely treat diseases that significantly affect the older people’s life quality in their later years, such as Alzheimer’s disease. Before determining treatment options, understanding the mechanisms involved in these diseases is required for pathological analysis. One of the materials: CRANAD-2, the neurogenic curcumin derivative, is thought to contribute to these studiesTorra2022VersatileNS ; bouzin2022melanin . IR images can detect CRANAD-2. High-resolution IR images would benefit for NIR nano imaging and further promote the development of correlational studies. Another representative example is the COVID-19 that has received much attention recently.

Considering the imbalance in development between different countries and geographies, it will create an unbalanced medical resource. Further, impacting the early detection to COVID-19 in developing countries is that costly lung imaging equipment is not available. Many physicians have attempted to use faster and cheaper X-ray images as a tool for disease diagnosisLukose2021OpticalTF ; CanalesFiscal2021COVID19CU . IR imaging plays a significant role in the field of engineering. IR cameras can be used to detect temperature changesjang2022automated in systems, which can help identify overheating components that may indicate potential failure or malfunction. In electrical engineering, this can be particularly useful in detecting overheating components that may cause power outages or other issues.

IR imaging can also be used to detect leaks in systems that utilize gases or liquids, such as pipelinesallred2021time , allowing engineers to quickly locate and repair these issues and avoid costly damage and... Additionally, IR imaging can be utilized to monitor the performance of systems and equipment, allowing engineers to identify potential problems before they become more serious and take action to prevent costly failures or downtimebutkevich2021photoactivatable... Part of the book series: Smart Innovation, Systems and Technologies ((SIST,volume 368)) Included in the following conference series: With the continuous advancement of information technology and the increasing demand for real-time infrared simulation, establishing a high-performance real-time infrared simulation system is the primary task of battlefield information support system and the key... 3D infrared scene simulation technology is widely used in military-oriented fields such as performance test and evaluation of weapon imaging system, infrared remote sensing and mapping.

In this chapter, driven by Artificial Intelligence (AI) technology, an infrared 3D scene modeling algorithm based on improved Convolutional Neural Network (CNN) is proposed. The calculation of material infrared image, the establishment of infrared image database and the real-time infrared scene modeling based on global 3D scene are preliminarily realized, and its modeling performance is simulated through experiments. The results show that after many iterations, the accuracy of the improved CNN is better than that of the traditional CNN, reaching more than 95%, and the error is also significantly reduced. The infrared image generated by this method is basically consistent with the real infrared image, which provides an effective way to prepare infrared reference map in real time. This is a preview of subscription content, log in via an institution to check access. Tax calculation will be finalised at checkout

A retrainable deep learning software allows to segment hearts and diagnose congenital heart disease in micro-computed tomography images of mouse models. CELTIC is a context-dependent deep learning model for in silico organelle localization that uses biological priors to enable the generalized and harmonized prediction of organelle fluorescence from label-free images. Grosjean et al. present a network-aware, self-supervised learning approach for screening neuronal activity dynamics. They demonstrate its applicability across a range of neural interventions. Identifying individual cells and tracking their movements over time are two critically important tasks in bioimage analysis.

We discuss a fresh wave of tools that push these techniques toward peak performance. Two new methodologies for automated cell tracking allow rapid analysis of cellular-scale behaviors across diverse samples with minimal errors.

Infrared Image Processing With Artificial Intelligence Progress And

People Also Search

This Special Issue Belongs To The Section “AI In Imaging“.

Despite These Advances, The Field Continues To Face Various Obstacles,

Use Of This Web Site Signifies Your Agreement To The

Compared To Visible Images, Infrared Images Can Be Better Tolerated

Then The Basic Components And Challenges Faced By Infrared Imaging