Multi-target tracking using Ramanujan Sum based DWT

Abstract

Wavelets are powerful set of tools used for Computer Vision. Ramanujan sums-based wavelet transform is relatively new in the field of signal processing. In this paper the Ramanujan Sum based DWT (Discrete wavelet transform) is implemented to tackle the problem of multitarget tracking using wavelets. A novel multi-target tracking algorithm is proposed based on DWT. The implementation is done for real time video sequences. The result is compared to Haar & Daubechies based DWT tracking algorithm and performance is better in the video sequences that have been evaluated on the Ramanujan sums-based DWT tracking algorithm. Also, as Ramanujan sums-based wavelet transform use integer coefficients and the normalization can be performed at a later stage, it would be easier to form hardware on this.