Hardware System Implementation for Human Detection using HOG and SVM Algorithm

TL;DR

This paper presents a hardware-based human detection system using HOG and SVM algorithms, achieving real-time performance with high accuracy and significantly reduced detection time compared to software implementations.

Contribution

The paper introduces a hardware architecture for human detection that accelerates computation and enables real-time application, which was not previously achieved with software alone.

Findings

Detection accuracy of 84.35%

Detection time reduced to 0.757 ms at 50MHz

System is 54 times faster than software implementation

Abstract

Human detection is a popular issue and has been widely used in many applications. However, including complexities in computation, leading to the human detection system implemented hardly in real-time applications. This paper presents the architecture of hardware, a human detection system that was simulated in the ModelSim tool. As a co-processor, this system was built to off-load to Central Processor Unit (CPU) and speed up the computation timing. The 130x66 RGB pixels of static input image attracted features and classify by using the Histogram of Oriented Gradient (HOG) algorithm and Support Vector Machine (SVM) algorithm, respectively. As a result, the accuracy rate of this system reaches 84.35 percent. And the timing for detection decreases to 0.757 ms at 50MHz frequency (54 times faster when this system was implemented in software by using the Matlab tool).