№2, 2022

REAL-TIME FACE DETECTION ON A RASPBERRY PI
Leyla G. Muradkhanli, Eshgin A. Mammadov

The article describes the implementation of different face detection algorithms to capture human faces from real-time video frames using a Raspberry PI microprocessor. This article examines this issue, proposes the implementation of two distinct real-time face detection algorithms, and presents a comprehensive architectural design. Used methods include Haar Cascades which is known as Viola-Jones algorithm, and Histogram of Oriented Gradients + Linear Support Vector Machines algorithm. The algorithms are implemented with the help of the OpenCV and Dlib libraries, and the Python programming language was used to build the face detection system. The OpenCV and Dlib libraries include a large number of built-in packages that assist with face detection and conduct face operations separately, resulting in reduced processing time and increased efficiency overall. The results confirm that both methods can detect faces in real time with acceptable accuracy and computation time but there are several differences. The Histogram of Oriented Gradients + Linear Support Vector Machines algorithm.method is much more preferable in terms of accuracy, but the image pyramid construction will be computationally demanding (pp.38-45).

Keywords:Face Detection, Raspberry PI, Histogram of Oriented Gradients, Support Vector Machines, Internet of Things
References
  • Xin, Zhang, Thomas, Gonnot. (2017). Real-Time Face Detection and Recognition in Complex Background. Journal of Signal and Information Processing , 8(2), 1-5.
  • Pablo, Tribaldos, Juan, Serrano-Cuerda. (2013). People Detection in Color and Infrared Video Using HOG and Linear SVM, 1-10. https://www.semanticscholar.org/paper/People-Detection-in-Color-and-Infrared-Video-Using-Tribaldos-Serrano-Cuerda/c933c4bef57be3585abb13bacb74aca29588a6ac
  • Jyotirmaya, Ijaradar, Jinjing, Xu. (2022) . A Cost-efficient Real-time Security Surveillance System Based on Facial Recognition Using Raspberry Pi and OpenCV.   Current Journal of Applied Science and Technology, 41(5), 1-12.
  • Klaus, Kollreider, Hartwi, Fronthaler. (2008). Real-Time Face Detection and Motion Analysis with Application in “Liveness” Assessment. IEEE Transactions on Information Forensics and Security, 2(3), 548 – 558.
  • Yi-Qing Wang. (2014). An Analysis of the Viola-Jones Face Detection Algorithm. Image Processing On Line, 4, 128-148.
  • Daniel Hefenbrock, Jason Oberg. (2010). Accelerating Viola-Jones Face Detection to FPGA-Level using GPUs. IEEE Xplore . Conference: Field-Programmable Custom Computing Machines (FCCM), 2010 18th IEEE Annual International Symposium on 11-17.
  • Haoxiang Li, Zhe Lin. (2015). A Convolutional Neural Network Cascade for Face Detection.  IEEE Explore, 1-10.
  • Li Cuimei1, Qi Zhiliang. (2017). Human face detection algorithm via Haar cascade classifier combined with three additional classifiers. International Conference on Electronic Measurement and Instruments, ICEMI, IEEE.
  • José Ignacio Rodríguez Molano. (2015). Internet of Things: A Prototype Architecture Using a Raspberry Pi. Lecture Notes in Business Information Processing, 224, 618-631.
  • Junjie Yan, Xuzong Zhang (2013). Face detection by structural models. Image and Vision Computing, Elsevier, 1-10.