TY - GEN
T1 - Computing parallel speeded-up robust features (P-SURF) via POSIX threads
AU - Zhang, Nan
PY - 2009
Y1 - 2009
N2 - Speeded-Up Robust Features (SURF), an image local feature extracting and describing method, finds and describes point correspondences between images with different viewing conditions. Despite the fact that it has recently been developed, SURF has already successfully found its applications in the area of computer vision, and was reported to be more appealing than the earlier Scale-Invariant Feature Transform (SIFT) in terms of robustness and performance. This paper presents a multi-threaded algorithm and its implementation that computes the same SURF. The algorithm parallelises several stages of computations in the original, sequential design. The main benefit brought about is the acceleration in computing the descriptor. Tests have been performed to show that the parallel SURF (P-SURF) generally shortened the computation time by a factor of 2 to 6 than the original, sequential method when running on multi-core processors.
AB - Speeded-Up Robust Features (SURF), an image local feature extracting and describing method, finds and describes point correspondences between images with different viewing conditions. Despite the fact that it has recently been developed, SURF has already successfully found its applications in the area of computer vision, and was reported to be more appealing than the earlier Scale-Invariant Feature Transform (SIFT) in terms of robustness and performance. This paper presents a multi-threaded algorithm and its implementation that computes the same SURF. The algorithm parallelises several stages of computations in the original, sequential design. The main benefit brought about is the acceleration in computing the descriptor. Tests have been performed to show that the parallel SURF (P-SURF) generally shortened the computation time by a factor of 2 to 6 than the original, sequential method when running on multi-core processors.
KW - Image processing
KW - Local feature extraction
KW - Machine intelligence
KW - Multi-threaded computing
KW - Parallel programming
UR - https://www.scopus.com/pages/publications/70350778331
U2 - 10.1007/978-3-642-04070-2_33
DO - 10.1007/978-3-642-04070-2_33
M3 - Conference Proceeding
AN - SCOPUS:70350778331
SN - 3642040691
SN - 9783642040696
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 287
EP - 296
BT - Emerging Intelligent Computing Technology and Applications - 5th International Conference on Intelligent Computing, ICIC 2009, Proceedings
T2 - 5th International Conference on Intelligent Computing, ICIC 2009
Y2 - 16 September 2009 through 19 September 2009
ER -