Construction of efficient detectors for character information recognition

TL;DR

This paper presents a universal approach for detecting various objects in video images, combining multiple feature extraction methods to improve recognition of characters and objects.

Contribution

It introduces a versatile detection framework that integrates different feature-based methods, enhancing recognition accuracy across diverse object types.

Findings

Compared efficiencies of Haar, LBP, and Census features

Constructed 11 detectors for railway carriage numbers

Achieved effective recognition of digits 0-9

Abstract

We have developed and tested in numerical experiments a universal approach to searching objects of a given type in captured video images (for example, people's faces, vehicles, special characters, numbers and letters, etc.). The novelty and versatility of this approach consists in a unique combination of the well-known methods ranging from creating detectors to making decisions independent of the type of recognition objects. The efficiencies of various types of basic features used for image coding, including the Haar features, the LBP features, and the modified Census transformation are compared. A combination of the modified methods is used for constructing 11 types of detectors of the number of railway carriages and for recognizing digits from zero to nine. The efficiency of the constructed detectors is studied.