Tejaswi with HTML5 Markup

Tejaswi Agarwal Graduate student @ UW-Madison

1. PHGRMS: A Parallel Hypergraph Based Root Mean Square Algorithm for Image Denoising

Guide: Dr. Rajesh Kanna, Vellore Institute of Technology, India

I presented this work at the 22nd ACM Symposium on High Performance Parallel and Distributed Computing, HPDC 2013, New York, USA . It also won the BEST POSTER Award at the conference among 26 selected posters.

I worked on this project with Saurabh Jha (final year undergraduate) on designing a parallel Salt and Pepper (SP) noise removal algorithm in a grey level digital image based on the Hypergraph Based Root Mean Square (HGRMS) approach. For SP noise removal, we reduce the original algorithm to a parallel model by introducing a cardinality matrix and an iteration factor, k, which helps us reduce the dependencies in the existing approach.We tested P-HGRMS using standard images from the Berkeley Segmentation dataset on NVIDIAs Compute Unified Device Architecture (CUDA) for noise identification and attenuation. P-HGRMS maintains the noise removal efficiency and outperforms its sequential counterpart by 6 to 18 times (6x - 18x) in computational efficiency. This work won the Best Poster award at HPDC 2013 in New York . The extended abstract and slides from my talk are here in the Research section.

The basic working of the algorithm can be explained by the figure below.



1. P-HGRMS maintains the noise removal efficiency of HGRMS algorithm as evident from PSNR values.
2. It outperforms HGRMS by 6 to 18 times (6x - 18x) in computational efficiency.


[1] K. Kannan, B. Rajesh Kanna, and C. Aravindan (2010), Root mean square filter for noisy images based on hypergraph model, Image and Vision Computing, 28 (9), 1329-1338, Elsevier, DOI: 10.1016/j.imavis.2010.01.013, 5- Year impact factor- 1.84

[2] Boukerrou and K. L. Kurz. Suppression of salt and pepper noise based on youden designs. Information Science, 110:217-235, 1998.

[3] R. H. Chan, C. W. Ho, and M. Nikolova. Salt and pepper noise removal by median type noise detectors and detail-preserving regularization. IEEE Transactions on Image Processing, 14(10):1479-1485, 2005.

[4] I. Frosio and N. A. Borghese. Human visual system modelling for real time salt and pepper noise removal. Biological and Artificial Intelligence Environments, pages 337-342, 2005

[5] H. S. L. H. C. Young and K. M. Poon. Modified CPI filter algorithm for removing salt and pepper noise in digital images in visual communications and image processing. In Proceedings of SPIE, pages 1439-1449, 1996.

[6] Rioulo. A spectral algorithm for removing salt and pepper noise from images. In Proceedings of Digital Signal Processing Workshop, Lone, pages 275-278, 1996.

[7] R. Dharmarajan, K. Kannan, "A hypergraph-based algorithm for image restoration from salt and pepper noise", International Journal of Electron Communication, AEU, 64 (2010) 1114-1122