Title:
|
RETRIEVAL AND CLASSIFICATION OF LEAF SHAPE BY SUPPORT VECTOR MACHINE USING BINARY DECISION TREE, PROBABILISTIC NEURAL NETWORK AND GENERIC FOURIER MOMENT TECHNIQUE: A COMPARATIVE STUDY |
Author(s):
|
Krishna Singh, Indra Gupta, Sangeeta Gupta |
ISBN:
|
978-972-8939-22-9 |
Editors:
|
Yingcai Xiao, Tomaz Amon and Piet Kommers |
Year:
|
2010 |
Edition:
|
Single |
Keywords:
|
SVM-BDT, PNN-PCNN, Fourier Moment |
Type:
|
Short Paper |
First Page:
|
412 |
Last Page:
|
417 |
Language:
|
English |
Cover:
|
|
Full Contents:
|
click to dowload
|
Paper Abstract:
|
The Leaf Recognition becomes important for Plant Classification. This paper presents classification of plants based on their leaf shape employing three different techniques, the Support Vector Machine with binary Decision Tree (SVM-BDT), Probabilistic Neural Network using Principle component analysis (PNN-PCNN) and Fourier moment for solving multiclass problems. In the proposed work these three techniques are used for comparing the performance of the classification of Leaves based on their classification accuracy. The proposed SVM based Binary Decision Tree architecture takes advantage of both the efficient computation of the decision tree architecture and the high classification accuracy of SVMs. This can lead to a dramatic improvement in recognition speed when addressing problems with large number of classes. Classification accuracy from all the three techniques are compared and it is observed that SVM-BDT performs better than PNN-PCNN and Fourier Moment technique. |
|
|
|
|