This paper develops a lifting scheme based on the recursive filtering of signals. The lifting scheme proposed operates with IIR filters and is different from the conventional lifting scheme. The lifting construction exploits prediction operator and update operator design based on the interpolation discrete splines in a spatial domain. The proposed scheme is based on interpolation and, as such, it involves only samples of signals and does not require any use of quadrature formulas and is more suitable for signal processing. Finally the performance of the algorithm is discussed by comparing the transform with the wavelet transform B9/7on denoise of two - dimension test images according to soft threshold of wavelet coefficient. The former outperforms slightly the latter. The advantages of lifting scheme lie in its flexibility in design and the lower cost in computation.