Segmentation in object-oriented encoding and multi-view images transmission

• Main Author: V. Yu. Tsviatkou
• Format: Article
• Language: Russian
• Published: Educational institution «Belarusian State University of Informatics and Radioelectronics» 2019-06-01
• Series: Doklady Belorusskogo gosudarstvennogo universiteta informatiki i radioèlektroniki
• Subjects: image segmentation; multi-view images; object-oriented image coding; transfer of multi-view images
• Online Access: https://doklady.bsuir.by/jour/article/view/1104

The urgency of the problem efficiency increasing of coding of the multi-view images for transmission over radio channels in mobile video surveillance systems and the need to develop methods, algorithms and specialized compression codecs with acceptable computational complexity are shown. The segmentation place in the object-oriented coding of multi-view images is determined. Expressions for the relative evaluation of the computational complexity of segmentation algorithms and object-oriented coding of multi-view images are proposed. It is shown that tree-wave segmentation and nodal quad-grids use reduces the computational complexity of object-oriented coding of multi-view images and, accordingly, reduces the delay in their transmission in mobile video surveillance systems.