Automatic synthesis of classifiers in FGPA

Abstract

In this paper an approach to automatic synthesis of classifiers in FPGAs based on AdaBoost algorithm is introduced. The various boosting methods are widely used for detection of objects in images. In this paper a novel approach to automatic synthesis of a classifier with respect to an efficient use of computing resources is proposed. The main purpose is to achieve an acceptable trade-off between speed and resources consumption of the designed classifier. The features, used in boosting algorithms, are based on Local Binary Patterns, which are more suitable for hardware implementation then traditionally used Haar-like features. Automatic synthesis of classifiers from trained data sets is also described. In the paper there are also described the results achieved.

Keywords

AdaBoost, FPGA, LBP, hardware synthesis.

Annotation

In this paper an approach to automatic synthesis of classifiers in FPGAs based on AdaBoost algorithm is introduced. The various boosting methods are widely used for detection of objects in images. In this paper a novel approach to automatic synthesis of a classifier with respect to an efficient use of computing resources is proposed. The main purpose is to achieve an acceptable trade-off between speed and resources consumption of the designed classifier. The features, used in boosting algorithms, are based on Local Binary Patterns, which are more suitable for hardware implementation then traditionally used Haar-like features. Automatic synthesis of classifiers from trained data sets is also described. In the paper there are also described the results achieved.

Published

2011

Pages

1–12

Proceedings

International Bata Conference for Ph.D. Students and Young Researchers

Conference

International Bata Conference for Ph.D. Student and Young Researchers

ISBN

978-80-7454-013-4

Publisher

Tomas Bata University in Zlín

Place

Zlín

BibTeX

@inproceedings{BUT76352,
  author="Filip {Kadlček} and Roman {Juránek} and Pavel {Zemčík}",
  title="Automatic synthesis of classifiers in FGPA",
  booktitle="International Bata Conference for Ph.D. Students and Young Researchers",
  year="2011",
  pages="1--12",
  publisher="Tomas Bata University in Zlín",
  address="Zlín",
  isbn="978-80-7454-013-4"
}

Projects

Advanced secured, reliable and adaptive IT, BUT, Vnitřní projekty VUT, FIT-S-11-1, start: 2011-01-01, end: 2013-12-31, completed
Security-Oriented Research in Information Technology, MŠMT, Institucionální prostředky SR ČR (např. VZ, VC), MSM0021630528, start: 2007-01-01, end: 2013-12-31, running

Departments

Department of Computer Systems (DCSY)