Weighted Ensemble Classifier for Plant Leaf Identification

R. Putri Ayu Pramesti, Yeni Herdiyeni, Anto Satriyo Nugroho

Abstract


Plant leaf identification using image can be constructed by ensemble classifier. Ensemble classifier executes classification of various features independently. This experiment utilized texture feature and geometry feature of plant leaf to find out which features are more powerful. Each classifier trained by specific feature produced different accuracy rate. To integrate ensemble classifier the results of the classification were weighted, so as the score obtained from better features contributed greater to the final results. Weighted classification results were combined to get the final result. The proposed method was evaluated using dataset comprises of 156 variety of plants with 4559 images. Weighting and combining classifier used in this study were Weighted Majority Vote (WMV) and Naïve Bayes Combination. Both of those method result showed better accuracy than using single classifier. The average accuracy of single classifier was 61.2% for geometry classifier and 70.3% for texture classifier, while WMV method was 77.8% and Naïve Bayes Combination was 94.6%. The calculation of classifier’s weight by using WMV method produces a weight value of 0.54 for texture feature classifier and 0.46 for geometry feature classifier.


Keywords


ensemble classifier, weighted classifier, naïve bayes combination, weighted majority vote

Full Text:

PDF


DOI: http://dx.doi.org/10.12928/telkomnika.v16i3.7615

Article Metrics

Abstract view : 305 times
PDF - 160 times

Refbacks



Copyright (c) 2019 Universitas Ahmad Dahlan

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

TELKOMNIKA Telecommunication, Computing, Electronics and Control
ISSN: 1693-6930, e-ISSN: 2302-9293
Universitas Ahmad Dahlan, 4th Campus, 9th Floor, LPPI Room
Jl. Ringroad Selatan, Kragilan, Tamanan, Banguntapan, Bantul, Yogyakarta, Indonesia 55191
Phone: +62 (274) 563515, 511830, 379418, 371120 ext. 4902, Fax: +62 274 564604

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

View TELKOMNIKA Stats