Infrared Dim Small Target Detection Using Median Filter

Implementation of infrared dim small target detection based on a 9x9 median filter window size, with algorithm efficiency analysis and code implementation considerations

Median Filter and Mean Filter

Median Filter and Mean Filter are two independently existing filters with no dependency relationship between them

Filtering Methodology of Adaptive Median Filter

Like traditional median filters, adaptive median filters employ a rectangular window region Sxy for processing. The key difference lies in the adaptive filter's ability to dynamically adjust (i.e., increase) the filter window size based on predefined conditions during operation. When the center pixel of the window is identified as noise, it gets replaced by the median value; otherwise, the original pixel value remains unchanged. The filter output subsequently replaces the value at coordinates (x,y), corresponding to the current window center position.

SAR Image Processing: Non-Statistical, Statistical, and Frequency-Domain Filtering Methods

Non-statistical filtering methods: Mean Filter, Median Filter, K-Nearest Neighbors Filter; Statistical filtering methods: Lee Local Statistics Filter, Maximum A Posteriori Filter; Frequency-domain approach primarily utilizes wavelet transform, notably the wavelet soft-thresholding method proposed by D.L. Doholo et al.