THE COMPUTER SYSTEM OF SHADOW LOCATING AND REMOVAL IN DIGITAL IMAGES
Науковий журнал «Радіоелектроніка, інформатика, управління»
Переглянути архів Інформація| Поле | Співвідношення | |
| ##plugins.schemas.marc.fields.042.name## |
dc |
|
| ##plugins.schemas.marc.fields.245.name## |
THE COMPUTER SYSTEM OF SHADOW LOCATING AND REMOVAL IN DIGITAL IMAGES |
|
| ##plugins.schemas.marc.fields.720.name## |
Serdiuk, М. Е.; Oles Honchar Dnipropetrovsk National University Berkut, V. G.; Oles Honchar Dnipropetrovsk National University |
|
| ##plugins.schemas.marc.fields.653.name## |
Shadow detection; shadow removal; image processing; Lab color space; segmentation; shadow model |
|
| ##plugins.schemas.marc.fields.520.name## |
<p>Context. Presence of shaded areas in digital images may cause problems in processes of objects recognition, tracking, classification. Thus shadows identification and removal are important pre-processing stages, which are used in many computer vision algorithms.</p><p>Objective. The work objective is to find effective methods for automatic shadow detection and removal and development of a software system that implements the appropriate image processing.</p><p>Method. In this paper we have discussed several approaches to the process of identifying and converting shaded areas, analyzed their applicability for different types of shadows. For shadow localization two methods are used: the method with the transition to the Lab color space and the method with preliminary removal of texture. For shadow removing four methods are used: additive method, method based on a mathematical model of shadows, method with the transition to the Lab-space, constant method. We have proposed some stages modifications to improve the practical results.</p><p>Results. Computer system with automatic detection and elimination of shadows, using the above methods, was developed. The choice of methods is carried out depending on the type of shaded areas. Some options of methods are available for configuration. The system was tested on series of images. We have analyzed quality and speed of the implemented methods. Image processing results by different methods are shown.</p><p>Conclusions. Implemented in the presented computer system methods of shadow locating and removing with the proposed modificationswork effectively in different types of images. Additional settings allow to improve the results in some difficult cases. The system can be used for image pre-processing in order to avoid errors in further work of the computer vision algorithms</p> |
|
| ##plugins.schemas.marc.fields.260.name## |
Zaporizhzhya National Technical University 2017-11-09 11:20:22 |
|
| ##plugins.schemas.marc.fields.856.name## |
application/pdf http://ric.zntu.edu.ua/article/view/112372 |
|
| ##plugins.schemas.marc.fields.786.name## |
Radio Electronics, Computer Science, Control; No 2 (2017): Radio Electronics, Computer Science, Control |
|
| ##plugins.schemas.marc.fields.546.name## |
uk |
|
| ##plugins.schemas.marc.fields.540.name## |
Copyright (c) 2017 М. Е. Serdiuk, V. G. Berkut |
|