Recognition of Leaf Images Based on Shape Features Using a Hypersphere Classifier

Lecture notes in computer sciencePublished 1 January 2005

Xiaofeng Wang, Ji‐Xiang Du, Guojun Zhang

DOI10.1007/11538059_10

Source URL

Citations83

SJR quartileQ2

SJR score0.35

SNIP0.55

TL;DR

A method of recognizing leaf images based on shape features using a hypersphere classifier using eight geometric features including rectangularity, circularity, eccentricity, etc, and seven moment invariants for classification is introduced.

Abstract

Recognizing plant leaves has so far been an important and difficult task. This paper introduces a method of recognizing leaf images based on shape features using a hypersphere classifier. Firstly, we apply image segmentation to the leaf images. Then we extract eight geometric features including rectangularity, circularity, eccentricity, etc, and seven moment invariants for classification. Finally we propose using a moving center hypersphere classifier to address these shape features. As a result there are more than 20 classes of plant leaves successfully classified. The average correct recognition rate is up to 92.2 percent.

Keywords

Earth and Planetary SciencesAgricultural and Biological Sciences

IRE Transactions on Information TheoryVisual pattern recognition by moment invariants

7,528 Citations1962Ming-Kuei Hu

It is shown that recognition of geometrical patterns and alphabetical characters independently of position, size and orientation can be accomplished and it is indicated that generalization is possible to include invariance with parallel projection.

IEEE Transactions on Pattern Analysis and Machine IntelligenceWatersheds in digital spaces: an efficient algorithm based on immersion simulations

5,468 Citations1991Luc Vincent, Pierre Soille

A fast and flexible algorithm for computing watersheds in digital gray-scale images is introduced, based on an immersion process analogy, which is reported to be faster than any other watershed algorithm.

Image Processing, Analysis and Machine Vision

3,719 Citations1993Milan Sonka, Václav Hlaváč +1 more

The digitized image and its properties are studied, including shape representation and description, and linear discrete image transforms, and texture analysis.

International Journal of Pattern Recognition and Artificial IntelligenceRADIAL BASIS PROBABILISTIC NEURAL NETWORKS: MODEL AND APPLICATION

424 Citations1999De-Shuang Huang

This paper proposes a new feedforward neural network model referred to as radial basis probabilistic neural network (RBPNN), which inherits the merits of the two old odels to a great extent, and avoids their defects in some ways.

Pattern RecognitionImproved moment invariants for shape discrimination

186 Citations1993Chien-Chang Chen

Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE<title>Improved moment invariants for shape discrimination</title>

176 Citations1993Chien-Chang Chen, Tung‐I Tsai

The theoretical improved moments computed based only on the shape boundary, some invariants derived from improved moments in variation to translation, rotation, and scaling are presented.

Neural ComputationA Neural Root Finder of Polynomials Based on Root Moments

111 Citations2004De-Shuang Huang, Horace H. S. Ip +1 more

This letter proposes a novel neural root finder based on the root moment method (RMM) to find the arbitrary roots (including complex ones) of arbitrary polynomials and compares the advantage of the RMM-based NRF over the previous root coefficient method based NRF and the traditional Muller and Laguerre methods.

IEEE Transactions on Neural NetworksZeroing Polynomials Using Modified Constrained Neural Network Approach

103 Citations2005Dongqing Huang, Horace H. S. Ip +2 more

This paper investigated different resulting constrained learning algorithms based on the variants of the error cost functions in the constrained BPN and derived a new modified CLA (MCLA), and found that the computational complexities of the CLA and the MCLAbased on the root-moment method (RMM) are the order of polynomial, and that the M CLA is simpler than the CLA.

Recognizing plant species by leaf shapes-a case study of the Acer family

83 Citations2002C. Im, Hirobumi Nishida +1 more

Experimental results indicate that structures of leaves and shapes of their components can be considered as landmarks to recognize the species of plants.

Automatic recognition of wild flowers

72 Citations2002Takeshi Saitoh, T. Kaneko

Which features are more effective for recognition are investigated and it is found that four features of flowers and two features of leaves can yield the best accuracy of 96%.