Session

Mechatronics, Robotics, and System Engineering

Description

In this paper we present several descriptors used for the task of Plant recognition based on the images of leaves. The set of descriptors include texture based descriptors, fractal descriptors as well as some of the state of the art descriptors for image retrieval and object recognition in images. The descriptors are generated from the leaf images taken from single leaves on homogenous background. The descriptors are then used for training classifiers from a dataset of leaf images. The comparison of the obtained results will be presented in this paper.

Keywords:

recognition, images, fractal, pixel

Session Chair

Gazmend Bajrami

Session Co-Chair

Kushtrim Dragusha

Proceedings Editor

Edmond Hajrizi

ISBN

978-9951-437-24-0

First Page

157

Last Page

159

Location

Durres, Albania

Start Date

1-11-2013 5:30 PM

End Date

1-11-2013 5:45 PM

DOI

10.33107/ubt-ic.2013.71

Share

COinS
 
Nov 1st, 5:30 PM Nov 1st, 5:45 PM

Comparison of local image descriptors for plant identification from leaf images

Durres, Albania

In this paper we present several descriptors used for the task of Plant recognition based on the images of leaves. The set of descriptors include texture based descriptors, fractal descriptors as well as some of the state of the art descriptors for image retrieval and object recognition in images. The descriptors are generated from the leaf images taken from single leaves on homogenous background. The descriptors are then used for training classifiers from a dataset of leaf images. The comparison of the obtained results will be presented in this paper.