Efficient 3-D parallel FIR filtering algorithm

In this paper we present an efficient 3-D digital filtering method based on a new 3-D parallel filtering algorithm. This method is suitable for high resolution / high speed 3-D image and video processing. The proposed 3-D parallel filtering algorithm is highly efficient due to the elimination of overlapping segments overhead in the block-filtering method, and the boundary conditions in parallel filtering applications. It also solves the problem of limited efficiency of direct FIR filtering when the system impulse response is large, and enhances the overall memory distribution of the parallel system by segmenting both the 3-D input data and the impulse response of the system into smaller independent subsections that can be simultaneously processed. Finally, the algorithm's performance is assessed and analyzed.

[1]  Pepe Siy,et al.  Parallel-decomposition algorithm for discrete computational problems and its application in developing an efficient discrete convolution algorithm , 1995 .

[2]  Alan Di Cenzo Transform lengths for correlation and convolution , 1987, IEEE Trans. Acoust. Speech Signal Process..

[3]  Said Boussakta,et al.  Parallelisation of the overlap-add block-filter algorithm for image processing , 2002, 2002 11th European Signal Processing Conference.

[4]  Jahangir Ahmad Satti Parallel beam mapping algorithm to compute radiation dose in three dimensional treatment planning , 1996, Proceedings Ninth IEEE Symposium on Computer-Based Medical Systems.

[5]  M.L.C. Hamilton The application of multiprocessor DSP to machine vision , 1995 .

[6]  M. Aziz,et al.  A hybrid parallel algorithm for digital image filtering applications , 2000, ICECS 2000. 7th IEEE International Conference on Electronics, Circuits and Systems (Cat. No.00EX445).

[7]  Desmond C. McLernon,et al.  High performance 2D parallel block-filtering system for real-time imaging applications using the Sharc ADSP21060 , 2003, Real Time Imaging.

[8]  C. Burrus,et al.  Fast one-dimensional digital convolution by multidimensional techniques , 1974 .

[9]  Zdenek Hanzalek Parallel processing: From applications to systems , 1997 .

[10]  Danny Crookes,et al.  Parallel architectures for image processing , 1998 .

[11]  Azriel Rosenfeld,et al.  Noise reduction in three-dimensional digital images , 1984, Pattern Recognit..

[12]  Charles M. Rader,et al.  Digital processing of signals , 1983 .

[13]  Erol Gelenbe,et al.  Multiprocessor Performance , 1990, SIGMETRICS Perform. Evaluation Rev..

[14]  Desmond C. McLernon,et al.  Parallelisation of the 1-D block filter algorithm to run on multiple DSPs , 2002, 9th International Conference on Electronics, Circuits and Systems.

[15]  Xinghai Li,et al.  Block size considerations for multidimensional convolution and correlation , 1992, IEEE Trans. Signal Process..