Faculty Advisor

Huang, Xinming

Abstract

This paper focuses on real-time pedestrian detection using the Histograms of Oriented Gradients (HOG) feature descriptor algorithm on a Field Programmable Gate Array. To achieve real- time pedestrian recognition on embedded systems, hardware architecture for HOG feature extraction is proposed. In order to reduce computational complexity toward efficient hardware architecture, this paper proposes several methods to simplify the computation of the HOG feature descriptor. The architecture is proposed on a Xilinx Zynq-7000 SoC using Verilog HDL to evaluate the real-time performance. This implementation processes image data at twice the pixel rate of similar software simulations and significantly reduces resource utilization while maintaining high detection accuracy.

Publisher

Worcester Polytechnic Institute

Date Accepted

April 2014

Major

Electrical and Computer Engineering

Project Type

Major Qualifying Project

Accessibility

Unrestricted

Advisor Department

Electrical and Computer Engineering

Share

COinS