DocumentCode :
2936863
Title :
Extracting Region with Holes with Edge Expanding Method
Author :
Han Hai ; Cheng Hui
Author_Institution :
Sch. of Math. & Comput. Sci., Jianghan Univ., Wuhan, China
Volume :
2
fYear :
2009
fDate :
18-20 Nov. 2009
Firstpage :
270
Lastpage :
273
Abstract :
In this paper, a method of region expanding evenly along edge is presented, meanwhile holes can be detected as byproducts. The complexity of the algorithm is O(n), where n is the quantity of pixels in the region. The source region is selected by 5 contiguous pixels with shape "+", of which the surrounding 4 pixels constitute the original edge. For every edge segment, a pixel out of the edge is checked whether it is similar to region pixels. The region is expanded by one pixel when a positive answer is given. During expanding period, if the pixel to be checked is already an edge pixel, the edge circle is divided into two parts - the one with more pixels remains as the surrounding edge, the other is the inner edge of a hole. Expanding process will stop if no change takes place in a repetition period.
Keywords :
computational complexity; feature extraction; edge expanding method; region extraction; Change detection algorithms; Computer science; Computer security; Data mining; Face recognition; Image color analysis; Image edge detection; Information security; Shape; Skin; circular linked list; detecting holes; edge expanding algorithm; region growing;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Multimedia Information Networking and Security, 2009. MINES '09. International Conference on
Conference_Location :
Hubei
Print_ISBN :
978-0-7695-3843-3
Electronic_ISBN :
978-1-4244-5068-8
Type :
conf
DOI :
10.1109/MINES.2009.221
Filename :
5370558
Link To Document :
بازگشت