Infrared Image Iterative Enhancement in Frequency Domain

This paper proposes an approach to infrared image enhancement involving iterative wrapping of the enhancement core within an image upscale/downscale loop. It highlights and suppresses distortion and blurs fine image details. This is the aim of the current study. Therefore, the isotropic pixel resolution of the digital image is chosen as an evaluation criterion. This criterion attunes the pre-processing result to the visual interpretation requirements of the infrared image. All processing workflow is performed inside the frequency domain, which is beneficial regarding computational burden. The actual infrared UAV image processing demonstrates an improvement in isotropic pixel resolution by a factor of 1.9–2.1, which is slightly superior to known single-pass algorithms. The proposed approach is flexible and modular and can be easily implemented in ground-based and onboard infrared imagery processing.