Presentation 2012-08-28
Tree Recognition using Leaf Images
Toshihiro IWATA, Takeshi SAITOH,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We propose a recognition method of trees using leaf images. We take two leaf images of surface and underside from a cut leaf on the white sheet. At this time, a marker is put on the background. At first, we extract the object region using a graph cuts algorithm. Secondly we define 25 features, and these features are fed to a nearest neighbor method. We collected 20 pairs of pictures from 50 trees in the fields around our campus. We found the underside image is better than the surface image. We obtained the recognition rates of 80.5% in the 1st candidate and 96.2% in the 5th candidates with 13 features. Moreover, we found the actual size improves 5% of recognition rate.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Tree recognition / leaf image / features / graph cuts
Paper # LOIS2012-28,IE2012-60,EMM2012-51
Date of Issue

Conference Information
Committee LOIS
Conference Date 2012/8/20(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Life Intelligence and Office Information Systems (LOIS)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Tree Recognition using Leaf Images
Sub Title (in English)
Keyword(1) Tree recognition
Keyword(2) leaf image
Keyword(3) features
Keyword(4) graph cuts
1st Author's Name Toshihiro IWATA
1st Author's Affiliation Dept. of System Design and Informatics, Kyushu Institute of Technology()
2nd Author's Name Takeshi SAITOH
2nd Author's Affiliation Dept. of System Design and Informatics, Kyushu Institute of Technology
Date 2012-08-28
Paper # LOIS2012-28,IE2012-60,EMM2012-51
Volume (vol) vol.112
Number (no) 188
Page pp.pp.-
#Pages 6
Date of Issue