Title :
Research on Segmentation Methods of Weed and Soil Background Under HSI Color Model
Author :
Li Dong-ming ; Wang Yuan-zhi ; Du Bo
Author_Institution :
Mech. & Electr. Coll., Agric. Univ. of Hebei, Baoding
Abstract :
In view of the complex background in the weed detection, the research used the H component under HSI the color model to divide the soil background. Considered the real-time operation, it used the automatic threshold value segmentation method, and carried on the suitable erosion and dilation, realized the weed area and the background area segmentation. The result indicated that it was possible of using above methods to obtain suited to segmentation of color space and the color characteristic component; therefore, it was practical and effective for threshold value parameter as the attribute image color information to segment weed from background area.
Keywords :
computer vision; image colour analysis; image segmentation; HSI color model; dilation; erosion; image segmentation; machine vision; soil; weed; Computer displays; Data mining; Educational institutions; Image analysis; Image processing; Image segmentation; Image storage; Machine vision; Soil; Spraying; dilation; erosion; image segmentation; machine vision; weed detection;
Conference_Titel :
Knowledge Discovery and Data Mining, 2009. WKDD 2009. Second International Workshop on
Conference_Location :
Moscow
Print_ISBN :
978-0-7695-3543-2
DOI :
10.1109/WKDD.2009.113