Akademik Veri Yönetim Sistemi
29th IEEE Conference on Signal Processing and Communications Applications (SIU), ELECTR NETWORK, 9 - 11 Haziran 2021
Thermal images have low contrast due to thermal radiation dispersion in the scene. It is generally difficult to separate objects from the background in these images. Thus, in IR image processing, it is a critical task to increase the details in the scene (Digital Detail Enhancement- DDE) and to reduce the noise, sharpening the edges and increasing the contrast in order to distinguish the targets in the scene properly. In this study, a suitable approach is proposed for improving thermal images by increasing the contrast and revealing details, low processing load and suitable for the hardware. In the proposed approach, thermal image is divided into low frequency and high frequency components using Gauss filter. Firstly, the adaptive double plateau histogram equalization method is applied to the low frequency component, followed by the Gamma Transform. In order to better reveal the details of the high frequency component, details were revealed by applying stepped gain. Then the images that hold the improved low frequencies and high frequencies are combined. In the final stage, noise removal is performed using Wiener Filter. When the experimental results are examined, it is shown that the proposed method performs well compared to other methods.