DocumentCode
2997480
Title
Edge detection of decayed wood image based on mathematical morphological double gradient algorithm
Author
Wang, Xueshun ; Qi, Dawei ; Li, Yuanxiang
fYear
2008
fDate
1-3 Sept. 2008
Firstpage
1226
Lastpage
1231
Abstract
Mathematical morphology is a new subject established based on rigorous mathematical theories. In the basis of set theory, mathematical morphology is used for image processing, analysing and comprehending. It is a powerful tool in the geometric morphological analysis and description. Based on the study of mathematical morphology, a new mathematical morphological double-gradient algorithm is proposed, and it is used in edge detection of decayed wood images. Structuring elements are chose appropriately, in order to suppress noises and be adapted to different edges of images. Mathematical morphological double gradient algorithm is constructed by weight adding combination of morphological operation. The results of simulation in decayed wood image processing demonstrate that the method performs better in noise-suppression and edge detection than conventional edge detection operations. Consequently, it provides a new method in processing decayed wood images.
Keywords
computational geometry; edge detection; gradient methods; image denoising; mathematical morphology; set theory; wood processing; decayed wood image processing; double gradient algorithm; edge detection; geometric morphological analysis; geometric morphological description; image noise suppression; mathematical morphology; set theory; Arithmetic; Automation; Forestry; Image edge detection; Image processing; Morphology; Multi-stage noise shaping; Set theory; Shape; Skeleton; Edge detection; Image processing; Mathematical morphology; wood nondestructive detection;
fLanguage
English
Publisher
ieee
Conference_Titel
Automation and Logistics, 2008. ICAL 2008. IEEE International Conference on
Conference_Location
Qingdao
Print_ISBN
978-1-4244-2502-0
Electronic_ISBN
978-1-4244-2503-7
Type
conf
DOI
10.1109/ICAL.2008.4636339
Filename
4636339
Link To Document