Separability Method for Homogeneous Leaves Using Spectroscopic Imagery and Machine Learning Algorithms

  • Bolanle Tolulope AbeEmail author
  • Jaco Jordaan
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11555)


Agathosma Buchu plants are a holistic healing system used as alternative medicine in dealing with numerous diseases and also for cooking oil production. There are two types namely, Betulina and Crenulata. The plants are difficult to separate if mixed up after harvest. Furthermore, the high rate of the plants’ cultivation poses challenges in separating them for specific functions. Hence, other identification methods are crucial. This paper presents an implementation of machine learning algorithms based on spectroscopic imagery properties for automatic recognition of the plants’ species. Image Local Polynomial Approximation method is used for the image processing to reduce classification error and dimensionality of classification challenges. To demonstrate the efficacies of the processed dataset, K-Nearest Neighbour, Naïve Bayes, Decision Tree, and Neural Network classifiers were used for the classification procedures in different data mining tools. The classifiers’ performances are valuable for decision-makers to consider tradeoffs in method accuracy versus method complexity.


Image Agathosma Separability Classification 



Our appreciation goes to Allen Harris the owner of the Buchu moon farm, near Cape Town, South Africa, for his support, taking us through his farm and giving us Buchu specimens used for this research. This work is based on the research supported wholly by the National Research Foundation of South Africa (Grant specific unique reference number (UID) 85745). The Grant holder acknowledges that opinions, findings and conclusions or recommendations expressed in any publication generated by the NRF supported research are that of the author(s), and that the NRF accepts no liability whatsoever in this regard.


  1. 1.
    Mavimbela, T., Viljoen, A., Vermaak, I.: Differentiating between Agathosma betulina and Agathosma crenulata–a quality control perspective. J. Appl. Res. Med. Aromat. Plants 1(1), e8–e14 (2014)Google Scholar
  2. 2.
    Van Wyk, B.E., Oudtshoorn, B.V., Gericke, N.: Medicinal Plants of South Africa. Briza Publications, Pretoria (1997)Google Scholar
  3. 3.
    Thring, T.S.A., Weitz, F.M.: Medicinal plant use in the Bredasdorp/Elim region of the southern overberg in the western cape province of South Africa. J. Ethnopharmacol. 103, 261–275 (2006)Google Scholar
  4. 4.
    Sandasi, M., et al.: Hyperspectral imaging and chemometric modeling of Echinacea-a novel approach in the quality control of herbal medicines. Molecules 19(9), 13104–13121 (2014)Google Scholar
  5. 5.
    Moolla, A., Viljoen, A.M.: Buchu – Agathosma betulina and Agathosma crenulata (Rutaceae): a review. Elsevier J. Ethnopharmacol. 119, 413–419 (2008)Google Scholar
  6. 6.
    Abe, B.T., Olugbara, O.O., Marwala, T.: Hyperspectral image classification using random forest and neural network. Lecture Notes in Engineering and Computer Science. In: Proceedings of the World Congress on Engineering and Computer Science, WCECS 2012, pp. 522–527, San Francisco, 24–26 October 2012Google Scholar
  7. 7.
    Van der Meer, F.D., van der Werff, H.M.A., van Ruitenbeek, F.J.A., Hecker, C.A., Bakker, W.H., Noomen, M.F., et al.: Multi- and hyperspectral geologic remote sensing: a review. Int. J. Appl. Earth Obs. Geoinf. 14, 112–128 (2012)Google Scholar
  8. 8.
    Landmann, T., et al.: Application of hyperspectral remote sensing for flower mapping in African savannas. Remote Sens. Environ. 166, 50–60 (2015)Google Scholar
  9. 9.
    Abe, B.T., Jordaan, J.A.: Identifying agathosma leaves using hyperspectral imagery and classification techniques. Lecture Notes in Engineering and Computer Science. In: Proceedings of the World Congress on Engineering and Computer Science, WCECS 2016, pp. 476–479, San Francisco, 19–21 October 2016Google Scholar
  10. 10.
    Sardogan, M., Tuncer, A., Ozen, Y.: Plant leaf disease detection and classification based on CNN with LVQ algorithm. In: 3rd International Conference on Computer Science and Engineering (UBMK), pp. 382–385, Sarajevo (2018)Google Scholar
  11. 11.
    Savltzky, A., Golay, M.J.E.: Smoothing and differentiation of data by simplified least squares procedures. Anal. Chem. 1084(36), 1627–1639 (1964)Google Scholar
  12. 12.
    Gorry, P.: General least-squares smoothing and differentiation by the convolution (Savitzky-Golay) method. Anal. Chem. 62(6), 570–573 (1990)Google Scholar
  13. 13.
    Katkovnik, V.: A new method for varying adaptive bandwidth selection. IEEE Trans. Signal Process. 47(9), 2567–2571 (1999)Google Scholar
  14. 14.
    Bialkowski, S.: Generalized digital smoothing filters made easy by matrix calculations. Anal. Chem. 61(11), 1308–1310 (1989)Google Scholar
  15. 15.
    Jordaan, J.A.: Fast and accurate spectral estimation algorithms for power system applications. Doctoral thesis, Tshwane University of Technology, South Africa (2006)Google Scholar
  16. 16.
    Larose, D.T.: Discovering Knowledge in Data an Introduction to Data Mining, pp. 90–106. Wiley Interscience, Hoboken (2005)Google Scholar
  17. 17.
    Okfalisa, I., Gazalba, M., Reza, N.G.I.: Comparative analysis of k-nearest neighbor and modified k-nearest neighbor algorithm for data classification. In: 2nd International Conferences on Information Technology. Information Systems and Electrical Engineering (ICITISEE), Yogyakarta, pp. 294–298 (2017)Google Scholar
  18. 18.
    Jahromi, A.H., Taheri, M.: A non-parametric mixture of Gaussian Naive Bayes classifiers based on local independent features. In: 2017 Artificial Intelligence and Signal Processing Conference (AISP), Shiraz, pp. 209–212 (2017)Google Scholar
  19. 19.
    Gui, J., Liu, T., Tao, D., Sun, Z., Tan, T.: Representative vector machines: a unified framework for classical classifiers. IEEE Trans. Cybern. 46(8), 1877–1888 (2016)Google Scholar
  20. 20.
    Trabelsi, A., Elouedi, Z., Lefevre, E.: Decision tree classifiers for evidential attribute values and class labels. Fuzzy Sets Syst. ISSN 0165-0114 (2018)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Department of Electrical EngineeringTshwane University of TechnologyeMalahleni CampusSouth Africa

Personalised recommendations