Extremum Median Filter Map Denoising Algorithm Based On Energy Function

The star maps acquired by the ground-based cameras are susceptible to the complex background of the starry sky and thus have high noise levels. In addition, the targets in star maps are similar to the noises due to their punctate shapes. As a result, the traditional image denoising method is not applicable to star maps. A new adaptive extremum median filtering denoising algorithm is put forward based on energy function, which can effectively remove the salt and pepper noise of the star maps and keep the small target information at the same time. This method employs a twice-check strategy to reduce the false detection ratio of noisy pixels and uses the improved adaptive median filter and the energy function model to recovery noise imagery. The simulated and real star map experiments show that, the Peak Signal to Noise Ratio (PSNR) is improved about 3 times and the Mean Squared Error (MSE) is reduced by 3.16x10(-5) in the terms of objective evaluations, the proposed method can effectively improve the denoising result and thus is applicable to star maps.