Title :
Fast Leaf Vein Extraction Using Hue and Intensity Information
Author :
Zheng, Xiaodong ; Wang, Xiaojie
Author_Institution :
Dep. of Comput. Sci. & Applic., Zhengzhou Inst. of Aeronaut. Ind. Manage., Zhengzhou, China
Abstract :
Leaf vein contains significant physiological information of plant. Leaf vein extraction is one of the key steps in plant modeling and plant recognition. Several methods have been presented on leaf vein extraction but all of them are limited in narrow application fields. In this paper a universal method is proposed using hue and intensity information. According to the color difference between leaf vein and mesophyll, leaves are classified as the contrasting color leaf and the concolorous leaf. Leaf vein of the contrasting color leaf can be extracted using only hue information. Leaf vein extraction of the concolorous leaf needs both hue and intensity information. The method has been programmed under Visual C++6.0 and many leaves from different plants have been tested. The results show a good applicability. Compared with other method, new method is simple, fast and universal. It can be embedded in the leaf feature extraction system and work in human-computer interaction way.
Keywords :
C++ language; biology computing; botany; feature extraction; image colour analysis; image segmentation; Visual C++6.0; color image segmentation; concolorous leaf; hue information; intensity information; leaf feature extraction; leaf vein extraction; mesophyll; physiological information; plant modeling; plant recognition; Aerospace industry; Application software; Computer industry; Computer science; Data mining; Feature extraction; Image color analysis; Image enhancement; Plants (biology); Veins;
Conference_Titel :
Information Engineering and Computer Science, 2009. ICIECS 2009. International Conference on
Conference_Location :
Wuhan
Print_ISBN :
978-1-4244-4994-1
DOI :
10.1109/ICIECS.2009.5364313