An approach to improve the classification accuracy of leaf images with dorsal and ventral sides by adding directionality features with statistical feature sets

Arun Kumar, Vinod Patidar, Deepak Khazanchi, G. Purohit, Poonam Saini

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Scopus citations

Abstract

The basic purpose of this work is to study statistical feature set obtained from digital leaf image with dorsal and ventral sides and to find the degree of classification accuracy for each dorsal and ventral leaf image dataset. Moreover, the effect of adding directional features to statistical feature set on the overall classification accuracy, is also investigated. The work also studies whether the ventral side of the digital leaf image can be a suitable alternative for classification of leaf image data set or not.

Original languageEnglish (US)
Title of host publicationAdvanced Computing and Communication Technologies - Proceedings of the 9th ICACCT, 2015
EditorsH.A. Nagarajaram, Ramesh K. Choudhary, Jyotsna Kumar Mandal, Nitin Auluck
PublisherSpringer Verlag
Pages89-97
Number of pages9
ISBN (Print)9789811010217
DOIs
StatePublished - 2016
Event9th International Conference on Advanced Computing and Communication Technologies, ICACCT 2015 - New Delhi, India
Duration: Nov 28 2015Nov 29 2015

Publication series

NameAdvances in Intelligent Systems and Computing
Volume452
ISSN (Print)2194-5357

Other

Other9th International Conference on Advanced Computing and Communication Technologies, ICACCT 2015
Country/TerritoryIndia
CityNew Delhi
Period11/28/1511/29/15

Keywords

  • Directionality
  • Dorsal and ventral sides
  • Leaf images
  • Statistical features

ASJC Scopus subject areas

  • Control and Systems Engineering
  • General Computer Science

Fingerprint

Dive into the research topics of 'An approach to improve the classification accuracy of leaf images with dorsal and ventral sides by adding directionality features with statistical feature sets'. Together they form a unique fingerprint.

Cite this