Color and Texture in Classification of Coconut
Siddesha S1, S K Niranjan2
1Siddesha S, Department of Computer Applications, JSS Science and Technology University, Mysuru, India.
2S K Niranjan, Department of Computer Applications, JSS Science and Technology University, Mysuru, India.
Manuscript received on 02 June 2019 | Revised Manuscript received on 10 June 2019 | Manuscript published on 30 June 2019 | PP: 1745-1750 | Volume-8 Issue-8, June 2019 | Retrieval Number: H6837068819 /19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Coconut is one of the commercial and versatile crops and is a part of day to day food in India. Classification of coconut is a process of separating the nuts based on its maturity. In this paper, we proposed a model for classifying the nuts by color and texture features. SVM is used as a classifier for classifying nuts into three separate classes, tender coconut (TC), mature coconut (MC) and Copra (CP). All these classes are used for different purposes and occasions. Color histogram and color moments are used as color features with wavelet, LBP, GLCM and Gabor features as texture features. Experimentation is conducted on a data set of 900 images with combination of color and texture features using SVM as classifier. In classification part, we used two SVM approaches One-Against-One and One-Against-All with four different kernel functions namely, linear, quadratic, polynomial and radial based. An accuracy of 99.07% is attained with the combination of color moments and Gabor using One-Against-All SVM for linear kernel function.
Keywords: Coconut, Color, Texture, SVM.
Scope of the Article: RFID Network and Applications.