DocumentCode :
682379
Title :
Research of target segmentation technology based on depth information
Author :
Duan Lian ; Piao Yan
Author_Institution :
Sch. of Electron. & Inf. Eng., Changchun Univ. of Sci. & Technol., Changchun, China
fYear :
2013
fDate :
23-24 Dec. 2013
Firstpage :
723
Lastpage :
726
Abstract :
For segmentation algorithm widespread over-segmentation problem, design a combination of depth information and color information approach. Firstly, the image depth information combined statistical area in which the fill color on the basis of information obtained initial segmentation results, then combine Graph Cuts algorithm to over-segmentation part, and finally get the exact target. Through a variety of algorithms can be combined advanced segmentation results. Simulation results show that the new algorithm can effectively inhibit the design over-segmentation and under-segmentation, more precise target.
Keywords :
graph theory; image colour analysis; image segmentation; color information approach; design over-segmentation; design under-segmentation; graph cuts algorithm; image depth information; statistical area; target segmentation technology; Algorithm design and analysis; Automation; Educational institutions; Image color analysis; Image segmentation; Instrumentation and measurement; Merging; Depth Information; Graph Cuts; Statistical Region Integration;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Instrumentation and Measurement, Sensor Network and Automation (IMSNA), 2013 2nd International Symposium on
Conference_Location :
Toronto, ON
Type :
conf
DOI :
10.1109/IMSNA.2013.6743378
Filename :
6743378
Link To Document :
بازگشت