Real-time HOG+SVM based object detection using SoC FPGA for a UHD video stream
Authors:
- 7 June 2022
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Abstract
Object detection is an essential component of many vision systems. For example, pedestrian detection is used in advanced driver assistance systems (ADAS) and advanced video surveillance systems (AVSS). Currently, most detectors (e.g. the YOLO - You Only Look Once - family) use deep convolutional neural networks. However, due to their high computational complexity, they are not able to process a very high-resolution video stream in real-time, especially within a limited energy budget. In this paper we present a hardware implementation of the well-known pedestrian detector with HOG (Histogram of Oriented Gradients) feature extraction and SVM (Support Vector Machine) classification. Our system running on $\text{AMD} \text{Xilinx\ Zynq\ UltraScale}+ \text{MPSoC}$ (Multiprocessor System on Chip) device allows real-time processing of 4K resolution (UHD - Ultra High Definition, $3840 \times 2160\ \text{pixels}$) video for 60 frames per second. The system is capable of detecting a pedestrian in a single scale. The results obtained confirm the high suitability of reprogrammable devices in the real-time implementation of embedded vision systems.
Keywords
Funding Information
- National Science Centre (2016/23/D/ST6/01389)
This publication has 19 references indexed in Scilit:
- Hardware efficient implementation of histograms of oriented gradients for pedestrian detection2017
- Embedded vision system for pedestrian detection based on HOG+SVM and use of motion information implemented in Zynq heterogeneous device2017
- Real-time pedestrian detection on a xilinx zynq using the HOG algorithm2015
- A scalable architecture for multi-class visual object detection2015
- A Novel On-Road Vehicle Detection Method Using $\pi$HOGIEEE Transactions on Intelligent Transportation Systems, 2015
- HOG Feature Extractor Hardware Accelerator for Real-Time Pedestrian Detection2014
- FPGA-Based Real-Time Pedestrian Detection on High-Resolution Images2013
- HoGG: Gabor and HoG-based human detection for surveillance in non-controlled environmentsNeurocomputing, 2013
- Architectural Study of HOG Feature Extraction Processor for Real-Time Object Detection2012
- Histograms of Oriented Gradients for Human Detection2005