INFORMATION TECHNOLOGIES
IMAGE PROCESSING TECHNIQUES BASED CROWD SIZE ESTIMATION
Mohanad Abdulhamid
AL-Hikma University, http://alhikma.edu.iq/
P.O. Box 10069, Baghdad, Iraq
E-mail: moh1hamid@yahoo.com
Lwanga Wanjira
University of Nairobi, http://uonbi.ac.ke/
P.O. Box 30197-00100, Nairobi, Kenya
E-mail: researcher12018@yahoo.com
Received February 26, 2020; peer reviewed March 30, 2020; accepted April 06, 2020
Abstract. Image processing algorithms are the basis for image computer analysis and machine Vision. Employing a theoretical foundation, image algebra, and powerful development tools, Visual C++, Visual Fortran, Visual Basic, and Visual Java, high-level and efficient computer vision techniques have been developed. This paper analyzes different image processing algorithms by classifying them in logical groups. In addition, specific methods are presented illustrating the application of such techniques to the real world images. In most cases more than one method is used. This allows a basis for comparison of different methods as advantageous features as well as negative characteristics of each technique is delineated. The main objective of this paper is to use image processing techniques to estimate the size of a crowd from a still photograph. The simulation results show that the different images have different efficiencies.
Keywords: image processing; crowd size estimation
UDC 004.932.2
RENSIT, 2020, 12(3):407-414.
DOI: 10.17725/rensit.2020.12.407.
Full-text electronic version of this article - web site http://en.rensit.ru/vypuski/article/348/12(3)407-414e.pdf