期刊名称：Lecture Notes in Engineering and Computer Science
印刷版ISSN：2078-0958
电子版ISSN：2078-0966
出版年度：2019
卷号：2240
页码：59-64
出版社：Newswood and International Association of Engineers
摘要：Segmentation is one of the significant requirements
of efficient computer vision applied in plant growth monitoring.
Existing segmentation techniques has their own merits, however
should be selected for a specific situation with respect to varying
plant environment. Consideration of segmentation in the context
of lettuce in hydroponics environment remain an open research.
In this paper, a lettuce plant segmentation by using thresholding
and super pixels is proposed, which can classify lettuce plant and
background from images taken at a smart farm hydroponics
setup. Lab color information of the image extracted from a
training image dataset undergo two-level thresholding and Kmeans
clustering thru superpixels to identify each pixel class.
Experimental testing results demonstrate an improved
performance in segmentation in terms sensitivity, precision, and
F1-score.
