Sugarcane Leaf Color Classification in Sa*b* Color Element Composition

R V Hari, Ginardi and Riyanarto, Sarno and Tri Adhi, Wijaya (2013) Sugarcane Leaf Color Classification in Sa*b* Color Element Composition. In: 2013 International Conference on Computer, Control, Informatics and Its Application (IC3INA), 19-20 November 2013, Jakarta-Indonesia.

[img]
Preview
Text
ic3ina 2013.pdf

Download (1MB) | Preview

Abstract

This research introduces a new approach in a color management for leaf color assessment. Currently, a RGB, HSV or LAB is used as a 3-D color space where color similarity can be evaluated by examining the closest distance of two colors,where each color is composed by three color elements. Saturation(S) element from HSV color space is chosen, combined with a* and b* elements from La*b* color space to represent a new color combination named Sa*b*. This combination has a strong representation of color spectrum which is important for color assessment in various lighting condition or various sensor. Atransformation from a RGB color space to Sa*b* is optimized with a Backpropagation Neural Netwok. The proposed method is suitable for leaf color assessment with various lighting conditions and different camera specifications. It gives an accuracy of color determination as high as 92 percent. This leaf color assessment is introduced to sugarcane growth monitoring as a part of sugarcane precision agriculture. This method can also be used to initialize a lef color chart for each variety of sugarcane. Keywords— sugarcane precision agriculture, leaf color assessment, color

Item Type: Conference or Workshop Item (Paper)
Subjects: Z Bibliography. Library Science. Information Resources > Z665 Library Science. Information Science
Z Bibliography. Library Science. Information Resources > ZA Information resources > ZA4050 Electronic information resources
Divisions: Fakultas Teknik > Teknik Informatika
Depositing User: Dwi Cahyono
Date Deposited: 15 Jun 2016 05:37
Last Modified: 15 Jun 2016 05:37
URI: http://repository.unitomo.ac.id/id/eprint/40

Actions (login required)

View Item View Item