Welcome to the InfoSci Platform
GPU Implementation of Image Convolution Using Sparse Model with Efficient Storage Format

GPU Implementation of Image Convolution Using Sparse Model with Efficient Storage Format

Saira Banu Jamal Mohammed, M. Rajasekhara Babu, Sumithra Sriram
Copyright: © 2018 |Volume: 10 |Issue: 1 |Pages: 17
ISSN: 1938-0259|EISSN: 1938-0267|EISBN13: 9781522543367|DOI: 10.4018/IJGHPC.2018010104
Cite Article Cite Article

MLA

Mohammed, Saira Banu Jamal, et al. "GPU Implementation of Image Convolution Using Sparse Model with Efficient Storage Format." IJGHPC vol.10, no.1 2018: pp.54-70. http://doi.org/10.4018/IJGHPC.2018010104

APA

Mohammed, S. B., Babu, M. R., & Sriram, S. (2018). GPU Implementation of Image Convolution Using Sparse Model with Efficient Storage Format. International Journal of Grid and High Performance Computing (IJGHPC), 10(1), 54-70. http://doi.org/10.4018/IJGHPC.2018010104

Chicago

Mohammed, Saira Banu Jamal, M. Rajasekhara Babu, and Sumithra Sriram. "GPU Implementation of Image Convolution Using Sparse Model with Efficient Storage Format," International Journal of Grid and High Performance Computing (IJGHPC) 10, no.1: 54-70. http://doi.org/10.4018/IJGHPC.2018010104

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

With the growth of data parallel computing, role of GPU computing in non-graphic applications such as image processing becomes a focus in research fields. Convolution is an integral operation in filtering, smoothing and edge detection. In this article, the process of convolution is realized as a sparse linear system and is solved using Sparse Matrix Vector Multiplication (SpMV). The Compressed Sparse Row (CSR) format of SPMV shows better CPU performance compared to normal convolution. To overcome the stalling of threads for short rows in the GPU implementation of CSR SpMV, a more efficient model is proposed, which uses the Adaptive-Compressed Row Storage (A-CSR) format to implement the same. Using CSR in the convolution process achieves a 1.45x and a 1.159x increase in speed compared to the normal convolution of image smoothing and edge detection operations, respectively. An average speedup of 2.05x is achieved for image smoothing technique and 1.58x for edge detection technique in GPU platform usig adaptive CSR format.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.