Leaf recognition using contour unwrapping and apex alignment with tuned random subspace method

Sruti Das Choudhury, Jin Gang Yu, Ashok K Samal

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

The variation in scale, translation and rotation pose the main challenges to automatic leaf recognition. This paper introduces an automatic leaf recognition method which uses generalised Procrustes analysis (GPA) to mutually align all leaf contours of each of the known classes with respect to scale, translation, rotation and reflection. A mean contour is computed as a representative of each known class. A test leaf is subjected to ordinary Procrustes analysis to be aligned with the mean contour with respect to scale, translation, rotation and reflection. However, experimental analyses show that in the cases where the leaf contours are significantly rotated with respect to each other, generalised Procrustes analysis fails to correctly align. To overcome this, we introduce a novel leaf apex detection algorithm based on Newton's divided method of interpolation and second order differentiation for critical point analysis. The 2-dimensional GPA-transformed contours are unwrapped by computing the distances between the contour points and the centre-of-mass of the contour starting from the leaf apex in an anticlockwise direction to generate a 1-dimensional distance signal. Principal component analysis is used for dimensionality reduction and linear discriminant analysis is used to achieve optimal class separability. The paper extends the use of random subspace method as an ensemble classifier in leaf recognition to exploit the high dimensionality of the feature space for improved identification by avoiding overlearning. Experimental analyses using two publicly available datasets demonstrate the effectiveness of the proposed method.

Original languageEnglish (US)
Pages (from-to)72-84
Number of pages13
JournalBiosystems Engineering
Volume170
DOIs
StatePublished - Jun 2018

Fingerprint

Differentiation (calculus)
leaves
Overlearning
Discriminant analysis
Newton-Raphson method
Principal component analysis
Interpolation
Discriminant Analysis
Classifiers
Principal Component Analysis
methodology
Recognition (Psychology)
method
alignment
discriminant analysis
interpolation
principal component analysis
analysis
testing
Direction compound

Keywords

  • Generalised Procrustes analysis
  • Leaf recognition
  • Random subspace method

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Food Science
  • Animal Science and Zoology
  • Agronomy and Crop Science
  • Soil Science

Cite this

Leaf recognition using contour unwrapping and apex alignment with tuned random subspace method. / Choudhury, Sruti Das; Yu, Jin Gang; Samal, Ashok K.

In: Biosystems Engineering, Vol. 170, 06.2018, p. 72-84.

Research output: Contribution to journalArticle

@article{2c7a806300bd41968962f0d195abfe88,
title = "Leaf recognition using contour unwrapping and apex alignment with tuned random subspace method",
abstract = "The variation in scale, translation and rotation pose the main challenges to automatic leaf recognition. This paper introduces an automatic leaf recognition method which uses generalised Procrustes analysis (GPA) to mutually align all leaf contours of each of the known classes with respect to scale, translation, rotation and reflection. A mean contour is computed as a representative of each known class. A test leaf is subjected to ordinary Procrustes analysis to be aligned with the mean contour with respect to scale, translation, rotation and reflection. However, experimental analyses show that in the cases where the leaf contours are significantly rotated with respect to each other, generalised Procrustes analysis fails to correctly align. To overcome this, we introduce a novel leaf apex detection algorithm based on Newton's divided method of interpolation and second order differentiation for critical point analysis. The 2-dimensional GPA-transformed contours are unwrapped by computing the distances between the contour points and the centre-of-mass of the contour starting from the leaf apex in an anticlockwise direction to generate a 1-dimensional distance signal. Principal component analysis is used for dimensionality reduction and linear discriminant analysis is used to achieve optimal class separability. The paper extends the use of random subspace method as an ensemble classifier in leaf recognition to exploit the high dimensionality of the feature space for improved identification by avoiding overlearning. Experimental analyses using two publicly available datasets demonstrate the effectiveness of the proposed method.",
keywords = "Generalised Procrustes analysis, Leaf recognition, Random subspace method",
author = "Choudhury, {Sruti Das} and Yu, {Jin Gang} and Samal, {Ashok K}",
year = "2018",
month = "6",
doi = "10.1016/j.biosystemseng.2018.04.001",
language = "English (US)",
volume = "170",
pages = "72--84",
journal = "Biosystems Engineering",
issn = "1537-5110",
publisher = "Academic Press Inc.",

}

TY - JOUR

T1 - Leaf recognition using contour unwrapping and apex alignment with tuned random subspace method

AU - Choudhury, Sruti Das

AU - Yu, Jin Gang

AU - Samal, Ashok K

PY - 2018/6

Y1 - 2018/6

N2 - The variation in scale, translation and rotation pose the main challenges to automatic leaf recognition. This paper introduces an automatic leaf recognition method which uses generalised Procrustes analysis (GPA) to mutually align all leaf contours of each of the known classes with respect to scale, translation, rotation and reflection. A mean contour is computed as a representative of each known class. A test leaf is subjected to ordinary Procrustes analysis to be aligned with the mean contour with respect to scale, translation, rotation and reflection. However, experimental analyses show that in the cases where the leaf contours are significantly rotated with respect to each other, generalised Procrustes analysis fails to correctly align. To overcome this, we introduce a novel leaf apex detection algorithm based on Newton's divided method of interpolation and second order differentiation for critical point analysis. The 2-dimensional GPA-transformed contours are unwrapped by computing the distances between the contour points and the centre-of-mass of the contour starting from the leaf apex in an anticlockwise direction to generate a 1-dimensional distance signal. Principal component analysis is used for dimensionality reduction and linear discriminant analysis is used to achieve optimal class separability. The paper extends the use of random subspace method as an ensemble classifier in leaf recognition to exploit the high dimensionality of the feature space for improved identification by avoiding overlearning. Experimental analyses using two publicly available datasets demonstrate the effectiveness of the proposed method.

AB - The variation in scale, translation and rotation pose the main challenges to automatic leaf recognition. This paper introduces an automatic leaf recognition method which uses generalised Procrustes analysis (GPA) to mutually align all leaf contours of each of the known classes with respect to scale, translation, rotation and reflection. A mean contour is computed as a representative of each known class. A test leaf is subjected to ordinary Procrustes analysis to be aligned with the mean contour with respect to scale, translation, rotation and reflection. However, experimental analyses show that in the cases where the leaf contours are significantly rotated with respect to each other, generalised Procrustes analysis fails to correctly align. To overcome this, we introduce a novel leaf apex detection algorithm based on Newton's divided method of interpolation and second order differentiation for critical point analysis. The 2-dimensional GPA-transformed contours are unwrapped by computing the distances between the contour points and the centre-of-mass of the contour starting from the leaf apex in an anticlockwise direction to generate a 1-dimensional distance signal. Principal component analysis is used for dimensionality reduction and linear discriminant analysis is used to achieve optimal class separability. The paper extends the use of random subspace method as an ensemble classifier in leaf recognition to exploit the high dimensionality of the feature space for improved identification by avoiding overlearning. Experimental analyses using two publicly available datasets demonstrate the effectiveness of the proposed method.

KW - Generalised Procrustes analysis

KW - Leaf recognition

KW - Random subspace method

UR - http://www.scopus.com/inward/record.url?scp=85046016220&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85046016220&partnerID=8YFLogxK

U2 - 10.1016/j.biosystemseng.2018.04.001

DO - 10.1016/j.biosystemseng.2018.04.001

M3 - Article

AN - SCOPUS:85046016220

VL - 170

SP - 72

EP - 84

JO - Biosystems Engineering

JF - Biosystems Engineering

SN - 1537-5110

ER -