Order statistics filters for color imaging

The design of multichannel filters, for example color imaging demands different objectives: noise suppression, chromaticity retention, and fine detail preservation. We propose a new class of multichannel filters, vector rank M-type K-nearest neighbor (VRMKNNF) filters to satisfy these objectives. The VRMKNNF filters are based on the combined RM-estimators with different influence functions. Also, we proposed using an adaptive non parametric approach to improve the performance of the filtering. The novel filter called adaptive multichannel non parametric VRMKNNF (AMN-VRMKNNF) filter has been investigated and presented in the paper. The VRMKNNF and AMN-VRMKNNF filters can successfully satisfy mentioned objectives, and also possess the robust and adaptive capability in color imaging applications. The simulation results demonstrate that the performance of proposed filters surpasses other existing filtering techniques. We also present the real-time implementation of proposed filters on the DSP TMS320C6711 demonstrating that they potentially provide a real-time solution to quality video transmission.