A Pipelined Multi-softcore Approach for the HOG Algorithm

• Tipo: Artigo em Livro de Atas de Conferência Internacional
• Data: 2016

Autores

Mascagni de Holanda, JAM (Autor) Outra; João M. P. Cardoso (Autor) FEUP; Marques, E (Autor) Outra

Ata de Conferência Internacional

Título: 2016 Conference on Design and Architectures for Signal and Image Processing (DASIP), Rennes, France, October 12-14, 2016

Páginas: 146-153

2016 Conference on Design and Architectures for Signal and Image Processing, DASIP 2016

12 October 2016 through 14 October 2016

Indexação

ISI Web of Knowledge - 1 Citação

Scopus - 1 Citação

Outras Informações

ID Authenticus: P-00M-FM5

DOI: 10.1109/DASIP.2016.7853811

Abstract (EN): This paper describes the mapping and the acceleration of an object detection algorithm on a multiprocessor system based on an FPGA. We use HOG ( Histogram of Oriented Gradients), one of the most popular algorithms for detection of different classes of objects and currently being used in smart embedded systems. The use of HOG on such systems requires efficient implementations in order to provide high performance possibly with low energy/power consumption budgets. Also, as variations and adaptations of this algorithm are needed to deal with different scenarios and classes of objects, programmability is required to allow greater development flexibility. In this paper we show our approach towards implementing the HOG algorithm into a multi-softcore Nios II based-system, bearing in mind high-performance and programmability issues. By applying sourceto-source transformations we obtain speedups of 19x and by using pipelined processing we reduce the algorithms execution time 49x. We also show that improving the hardware with acceleration units can result in speedups of 72.4x compared to the embedded baseline application.

Idioma: Inglês

Tipo (Avaliação Docente): Científica

Nº de páginas: 8