A class of circularly-symmetric CNN spatial linear filters

Radu Matei, Liviu Goraş
2006 Facta universitatis - series Electronics and Energetics  
This paper proposes a simple and efficient method for designing a class of circularly-symmetric spatial linear filters implemented on cellular neural networks. The design method relies on a so-called 1-D prototype filter, with desired characteristics and on a 1-D to 2-D spatial frequency transformation. Several design examples are given, for 2-D low-pass and band-pass filters (both of FIR and IIR type) with imposed cut-off or peak frequency and a specified selectivity. Finally, simulation
more » ... y, simulation results are provided, on a real grayscale biomedical image.
doi:10.2298/fuee0602299m fatcat:nis56t4ncfhk5poxpmqlgjvm2y