A New Algorithm of Speckle Filtering using Stochastic Distances
This is an incremental improvement for SAR image processing, addressing noise reduction in homogeneous regions.
The paper tackles speckle noise reduction in SAR imagery by introducing a filter based on stochastic distances and goodness-of-fit tests, which selects only statistically similar pixels for averaging. It shows this method outperforms Lee's filter in metrics like equivalent number of looks and edge preservation, with concrete improvements in quality indices.
This paper presents a new approach for filter design based on stochastic distances and tests between distributions. A window is defined around each pixel, overlapping samples are compared and only those which pass a goodness-of-fit test are used to compute the filtered value. The technique is applied to intensity SAR data with homogeneous regions using the Gamma model. The proposal is compared with the Lee's filter using a protocol based on Monte Carlo. Among the criteria used to quantify the quality of filters, we employ the equivalent number of looks, line and edge preservation. Moreover, we also assessed the filters by the Universal Image Quality Index and the Pearson's correlation on edges regions.