Research on multiresolution image enhancement methods of spot shaped targets in airborne SAR

Airborne Synthetic Aperture Radar (SAR), which is not only able to work under all day and all weather condition, but also more flexible and short period observation than that of space borne SAR, is fit for detecting and tracking suspicious targets which includes spot-shaped targets and block-shaped targets. There are inevitably some speckle noises in an airborne SAR due to its imaging mechanism, and the noise adversely affects target detection and identification, so enhancement is necessary before target´s detection. Suspicious targets in SAR images usually show bright spot-shaped features, common de-noising methods reduce noises while wipe some suspicious targets at the same time. Here we chose use Frost filter based on the comparison of some spatial filtering methods. Airborne SAR images usually have great sizes and like a strip, for the size of the image on the flight path direction is much larger than the scanning width of SAR. Enhancement directly on the original image can not meet the real-time or near real-time requirements of practical application. In this paper, we uses a multi-resolution approach based on the Gaussian pyramid firstly, and then use Frost filter. We demonstrate through several examples the effectiveness of the algorithm in maintaining the target features and accelerating the processing of enhancement.