Title :
Active Contour Model (Snake) for Wood Cell Contour Extraction
Author :
Diao, Hongzhi ; Cao, Jun ; Li, Guiying
Author_Institution :
Northeast Forestry Univ., Harbin
Abstract :
In the view of the situation that certain quantities of cells adhere, overlap and gather when the outline edge of cells is extracted during the research on forming characteristic of wood cells, this paper puts forward the improved two-step inflation model of snake with which outline of cells is extracted. At the same time, the greedy optimized algorithm is applied to reduce the running time of the procedure. This method finishes the segmentation to the cell cavity of wood at first step, and then to the cell wall of the wood, which enhances the effect of the exterior strength and the precision of capturing the cell outline, and also helps to reduce the impact on the possibility of trapping in the local minimum point of the object energy function
Keywords :
edge detection; feature extraction; forestry; greedy algorithms; optimisation; active contour model; cell segmentation; greedy optimized algorithm; snake model; wood cell contour extraction; Active contours; Automation; Decision support systems; Energy capture; Forestry; Greedy algorithms; Intelligent control; cell segmentation; greedy algorithm; snake model; wood cell;
Conference_Titel :
Intelligent Control and Automation, 2006. WCICA 2006. The Sixth World Congress on
Conference_Location :
Dalian
Print_ISBN :
1-4244-0332-4
DOI :
10.1109/WCICA.2006.1714042