Performance Evaluation of Infrared Image Enhancement Techniques

TL;DR

This paper provides a comprehensive survey of infrared image enhancement techniques, covering IR radiation types, devices, datasets, and various enhancement methods including spatial, frequency-domain, and deep learning approaches.

Contribution

It offers an up-to-date, detailed overview of IR image enhancement methods and categorizes existing techniques across different domains and technologies.

Findings

Summarizes IR radiation types and devices.

Reviews spatial and frequency-domain enhancement techniques.

Includes recent deep learning-based methods.

Abstract

Infrared (IR) images are widely used in many fields such as medical imaging, object tracking, astronomy and military purposes for securing borders. Infrared images can be captured day or night based on the type of capturing device. The capturing devices use electromagnetic radiation with longer wavelengths. There are several types of IR radiation based on the range of wavelength and corresponding frequency. Due to noising and other artifacts, IR images are not clearly visible. In this paper, we present a complete up-todate survey on IR imaging enhancement techniques. The survey includes IR radiation types and devices and existing IR datasets. The survey covers spatial enhancement techniques, frequency-domain based enhancement techniques and Deep learning-based techniques.