Real-time moving human detection using HOG and Fourier descriptor based on CUDA implementation
- PDF / 1,316,871 Bytes
- 16 Pages / 595.276 x 790.866 pts Page_size
- 58 Downloads / 200 Views
SPECIAL ISSUE PAPER
Real‑time moving human detection using HOG and Fourier descriptor based on CUDA implementation Haythem Bahri1 · Marwa Chouchene1 · Fatma Ezahra Sayadi1 · Mohamed Atri1,2 Received: 3 June 2019 / Accepted: 29 November 2019 © Springer-Verlag GmbH Germany, part of Springer Nature 2019
Abstract Real-time applications of image and video processing algorithms have seen explosive growth in number and complexity over the past decade driven by consumer, scientific and defense applications exploiting inexpensive digital video cameras and networked computing device. This growth has opened up different alternatives to greatly enhance the surveillance capabilities using new architectures and parallelization strategies developed due to the increased accessibility of multicore, multi-threaded processors along with general purpose graphics processing units (GPUs). In this paper, we present a new implementation of a moving human detection algorithm on GPU based on the programming language CUDA. In our approach, the moving object is extracted by background subtraction based on the GMM (Gaussian Mixture Model) on GPU. Then, two complementary features are extracted for moving object classification. They are contour-based description: FD or Fourier Descriptor and region-based description: HOG or Histogram of Oriented Gradient. Both descriptors will then be effectively integrated to SVM (Support Vector Machine), which is able to provide the posterior probability, to achieve better performance. The implementation of such algorithm on a GPU allows a great performance in terms of execution time since it is 19 times faster than that on a CPU. Experimental results show also that the proposed approach outperforms some existing techniques and can detect pedestrians in real-time effectively. Keywords GPU · CPU · CUDA · HOG · Fourier descriptor · GMM · Moving human detection · Real-time
1 Introduction In recent years, the security concerns have grown tremendously. During these decades, it is very remarkable that research in the field of security is increasingly sought after in different areas, such as borders where buffer zones are of utmost importance. That is why Video surveillance systems have been evolving significantly over the years and are becoming a vital tool for many organizations for safety and security applications. Two key factors brought on the popular use of such digital systems are a good accuracy and a detection speed meeting real-time requirements. Despite various approaches to satisfy such demands, none has really * Haythem Bahri [email protected] 1
Laboratory of Electronics and Microelectronics, Faculty of Sciences of Monastir, University of Monastir, 5000 Monastir, Tunisia
College of Computer Science, King Khalid University, Abha, Saudi Arabia
2
brought definitive solutions, because, according to Shashua [1], detection is confronted to a very great variability in the form of a person. Today, the trend is to combine artificial vision methods and learning methods such as neural networks
Data Loading...