Abstract
The most common technique used for image processing applications is ‘The wavelet transformation’. The Discrete Wavelet Transform (DWT) keeps the time as well as frequency information depend on a multi resolution analysis structure, where the other classical transforms like Fast Fourier Transform (FFT), Discrete Cosine Transform (DCT) will not do that. Because of this feature, the quality of the repaired image is improved when comparing to the other transforms. To implement the DWT on a real time codec, a fast device needs to be targeted. While comparing with the other implementation such as PCs, ARM processors, DSPs etc, Field Programmable Gate Array (FPGA) implementation of DWT had better processing speed and costs were vey less. A Fast Architecture based DWT using Kogge Stone Adder is proposed in this paper where the coefficients of lifting scheme are calculated by using Shift adder and Kogge Stone Adder where other techniques used multiplier. The most important intention of the suggested technique is to use minimum calculation and limited memory. The simulation of the suggested design is dole out on the Xilinx 14.1 style tool and also the performance is evaluated and compared with the present architectures.
Highlights
The suggested technique (Discrete Wavelet Transform) is playing a major role in signal analysis, computer vision, object detection, image confining and video confining standard fields [20]
The reproduction of the suggested design in implemented in Xilinx 14.1 design tool and it has been assessed by the resultants such as synthesis report and Field Programmable Gate Array (FPGA) device utilization summary
Looking at the above advantages, we can conclude that the suggested Kogge stone adder based multiplier consumes less power and it can be employed in the lifting scheme based discrete wavelet transform (DWT) architecture
Summary
The suggested technique (Discrete Wavelet Transform) is playing a major role in signal analysis, computer vision, object detection, image confining and video confining standard fields [20]. To handle the need of real time processing, various VLSI architectures for the 2-D DWT is suggested. Lifting scheme is used to avoid these kinds of problems Since this scheme uses basic convolution equations, the number of multipliers needed will be very less. DWT architecture is basically based on lifting scheme. Due to the less calculation problems, the lifting scheme of computation of DWT is preferred more than convolution-based scheme in most cases. Looking at the designing time, Kogge-stone adder is the fastest adder This is being a default choice for the industry as it is a high-performance adder.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
