Identification System for Moving Objects Based on Parallel Shift Technology

Identification System for Moving Objects Based on Parallel Shift Technology

Sergey Yuzhakov (Haisyn Department of the SFS General Directorate in Vinnytsia Region, Ukraine) and Stepan Mykolayovych Bilan (State University of Infrastructure and Technology, Ukraine)
DOI: 10.4018/978-1-7998-1290-6.ch015

Abstract

There are tasks of automatic identification of the moving stock of the railway, one of which is the automatic identification of rail cars cars by their number plates. Different organizational, legal, moral and ethical, technical, and programmatic methods of automated identification are used to solve this problem. At present little attention is paid to the development of means of automatic identification of moving objects, which would be possible regardless of the orientation and shape of the figure, especially if it concerns the recognition of freely oriented images of number plates. Therefore, many new methods for recognizing of number plates are developing. In the chapter, the system of identification of objects by their number plates in real time is considered. On moving objects (moving stock of a railway), an identifier image is drawn, which is an ordered set of characters. As a rule, these are numbers. But there may be other characters. The work also discusses the method of identification images of number plates with a high percentage of noise.
Chapter Preview
Top

Principles Of Implementation Of Parallel Shift Technology

The chapter focuses on the use of parallel shift technology (PST) to obtain the main characteristic features when describing certain objects (Belan at al. 2013a, Belan at al. 2013b, Bilan at al. 2018). This technology is based on the interaction of the original object, which needs to be described and its copy, which shifts in some space. The shift of elements occurs in parallel. The number of directions of the shift characterizes the accuracy of the object description.

The area of intersection of the initial image and its copy should be calculated for each iteration during a displacement. A sequence of values of the intersection area is formed for each selected direction. The number of such sets is determined by the number of shift directions. The displacement length is determined by the maximum distance between the extreme points of the initial image, which are located on one straight, corresponding direction of displacement. Each sequence of intersection areas can be turned into a function. Such functions are called the function of the area of intersection (FAI) and can be presented analytically, tabularly or graphically.

The method can be used for all tasks of identifying both stationary and moving objects. It is especially effective for the identification of moving objects, since the image shift occurs in the process of movement of the identification object itself, on which the image of the identifier is depicted.

Complete Chapter List

Search this Book:
Reset