Automated classification of tropical shrub species: A hybrid of leaf shape and machine learning approach / Miraemiliana Murat

Miraemiliana , Murat (2018) Automated classification of tropical shrub species: A hybrid of leaf shape and machine learning approach / Miraemiliana Murat. Masters thesis, University of Malaya.

[img] PDF (The Candidate's Agreement)
Restricted to Repository staff only

Download (174Kb)
    PDF (Thesis M.A.)
    Download (4Mb) | Preview


      Plants play an important role in foodstuff, medicine, industry, and environmental protection. The plant recognition is very crucial in some applications, including conservation of endangered species and rehabilitation of lands after mining activities. But, it is a challenging task to identify plant species because it requires specialised knowledge. Therefore, developing an automated classification system for plant species is necessary and valuable since it can help specialists as well as the public in identifying plant species easily. In this study, shape descriptors are applied on the myDAUN dataset that contains 45 tropical shrub species, which are collected from the University of Malaya (UM), Malaysia. Four types of shape descriptors are used in this study namely morphological shape descriptors (MSD), histogram of oriented gradients (HOG), Hu invariant moments (Hu) and Zernike moments (ZM). Single descriptor, as well as the combination of hybrid descriptors are tested and compared. The tropical shrub species are classified using six different classifiers, which are artificial neural network (ANN), random forest (RF), support vector machine (SVM), k-nearest neighbour (k-NN), linear discriminant analysis (LDA) and directed acyclic graph multiclass least squares twin support vector machine (DAG MLSTSVM). In addition, three types of feature selection methods are tested in the myDAUN dataset, namely Relief, Correlation-based feature selection (CFS) and Pearson’s coefficient correlation (PCC). The well-known Flavia dataset and Swedish Leaf dataset are used as the validation dataset on the proposed methods. The results showed that the hybrid of all descriptors of ANN outperformed the other classifiers with an average classification accuracy of 98.23% for myDAUN dataset, 95.25% for Flavia dataset and 99.89% for Swedish Leaf dataset. In addition, the Relief feature selection method achieved the highest classification accuracy of 98.13% after 80 (or 60%) of the original features are reduced, from 133 to 53 descriptors in myDAUN dataset with the reduction in computational time. Subsequently, the hybridisation of four descriptors gave the best results compared to others. It is proven that the combination of MSD and HOG are good enough for tropical shrubs species classification. Hu and ZM descriptors also improved the accuracy in tropical shrubs species classification in terms of invariant to translation, rotation and scale. ANN outperformed the others for tropical shrub species classification in this study. Feature selection methods can be used in the classification of tropical shrub species, as the comparable results could be obtained with the reduced descriptors while reducing in computational time and cost.

      Item Type: Thesis (Masters)
      Additional Information: Dissertation (M.A.) – Faculty of Science, University of Malaya, 2018.
      Uncontrolled Keywords: Shape descriptors; Feature extraction; Species identification; Machine learning; Artificial neural network
      Subjects: Q Science > Q Science (General)
      Q Science > QH Natural history > QH301 Biology
      Divisions: Faculty of Science
      Depositing User: Mr Mohd Safri Tahir
      Date Deposited: 19 Aug 2020 08:04
      Last Modified: 19 Aug 2020 08:04

      Actions (For repository staff only : Login required)

      View Item