文献类型： 会议论文

第一作者： L. Journaux

作者： L. Journaux ¹ ; M.F. Destain ² ; F. Cointault ¹ ; J. Miteran ³ ; A. Piron ² ;

作者机构： 1.Enesad, Engineering Sciences, 26 Bd Dr Petitjean, BP 87999, 21079 Dijon Cedex, France

2.FUSAGX, Unite de Mecanique et Construction, 2passage des deporte, 5030 Gembloux, Belgium

3.Universite de Bourgogne, Le2i, Avenue Alain Savory, BP 47870, 21078 Dijon Cedex, France

关键词： texture classification;motion descriptors;dimensionality reduction;leaf roughness

会议名称： European Conference on Precision Agriculture

主办单位：

页码： 213-220

摘要： In the context of precision spraying research, this article explores the capacity and the performance of some combinations of pattern recognition and computer vision techniques applied to plant leaf roughness analysis. The techniques merge feature extraction, linear and nonlinear dimensionality reduction techniques and several kinds of classification methods. The performances are evaluated and compare in terms of the classification error. We concluded of the well performance of leaf roughness characterisation through the combination of Generalized Fourier Descriptors for feature extraction, SVM for classification method and Kernel-Discriminant for dimensionality reduction.

分类号： S127-532