Абстрактный

ASIC Implementation of Multiplexer Based DAA

D.Srimathi, P.Santhini, M.Malarvizhi, B.SahayaJenila

In Digital Image Processing Point, Line and Edge detection are performed through software approach. The proposed Architecture performs these operations through hardware approach using Distributed Arithmetic. Distributed arithmetic (DA) has been widely used to implement inner product computations with fixed inputs. Conventional ROM-based DA suffers from large ROM requirements. To reduce the memory requirements, Adder based DA uses pre-defined structure for computation. But both the methods are suitable only if at least one input is constant. This project aims to implement a new Distributed Arithmetic Architecture for point detection, line detection and edge detection in DIP when both the inputs are variable. The new architecture is termed as Multiplexer based Distributed Arithmetic (MUX based DA). The proposed architecture takes the advantage of Multiplexer and DA for inner product computations when both the inputs are variable. In addition it reduces ROM requirement and complexity in constructing Adder based architecture for higher order inputs. Here, the performance of proposed Architecture with ROM based DA, Adder based DA and with multiplier based implementation are compared. The MUX based DA reduces power up to 81% and needs 40% of area as compared with multiplier based implementation.