Design sensitivity analyses of two-dimentional recursive band-pass and band-stop digital filters with an application in image processing

Two-dimensional variable recursive digital Band-pass and Band-stop filters are applied in signal processing and pro-imagining process, as well as communication systems where the frequency domain characteristics of digital filters are required to be adjustable. The main objective of this thesis has been to propose a new method of designing 2-D recursive Band-Pass and Band-Stop digital filters with variable characteristics. From the identical analog 1-D second order Butterworth Low-Pass analog ladder network, 2-D Band-Pass and Band-Stop digital filters can be obtained through the application of Low-Pass to Band-Pass/Band-Stop transformation and double generalized bilinear transformations. The denominators of these filters transfer functions are verified for VSHP. Sensitivity analyses are performed by varying the coefficients of the double generalized bilinear transformation such as k 1 , k 2 , a 1 , a 2 , b 1 and b 2 with respect to center frequency '} o ' and Bandwidth 'B' on the resulted 2-D Band-Pass and Band-Stop filter, which are obtained by varying the coefficients of the double generalized bilinear transformation in the specific ranges in order to maintain the stability of the filter of the Band-Pass and Band-Stop transfer function respectively.