Abstract
Recently, a wavelet-based dynamic range compression technique has been introduced to improve the visual quality of digital images captured in scenes with high dynamic range and uneven lighting conditions. This method enhances image visibility while maintaining important details such as local contrast and natural tonal appearance. Because of these advantages, the algorithm is particularly useful for aerial image processing applications, including image analysis tasks related to defense and security.
The proposed enhancement technique can also be applied to video transmission systems used in aviation, where improved image clarity is important for safety. In this project, an updated version of the algorithm is presented. The improved method enhances aerial images in such a way that the processed images can provide clearer visual information than direct observation in certain situations. Experimental results obtained from multiple aerial images demonstrate that the algorithm provides reliable performance and produces high-quality enhanced images.
Images captured from aircraft, satellites, or spacecraft often experience degradation due to atmospheric conditions surrounding the Earth. Environmental factors such as haze, fog, clouds, rain, or dust can reduce image clarity and visibility. In severe conditions, the captured images may appear extremely unclear, sometimes even approaching near-zero visibility for the human eye. Despite this limitation, valuable information may still be present in these images, which can be extracted through image enhancement techniques.
In many cases, the images recorded by imaging devices do not accurately represent the real-world scene because of the limited dynamic range of sensors. Real-world environments often contain a very wide range of brightness levels, while cameras and sensors can capture only a limited range. This limitation results in images with poor local contrast and reduced detail visibility.
The Human Visual System (HVS) naturally adapts to such conditions by compressing dynamic range and adjusting locally to different regions of the scene. However, in challenging environments such as foggy, rainy, or snowy conditions, both human observation and captured images may suffer from extremely low contrast due to the narrow dynamic range of the scene. Therefore, effective image enhancement methods are required to improve the visibility and interpretability of aerial imagery captured under such difficult conditions.