Linear GP with Redundancy-removed Recombination for Synthesis of Image Feature Extraction Programs

元データ 2008-12-10 社団法人情報処理学会

概要

We describe an evolutionary synthesis of feature extraction program for object recognition. The evolutionary synthesis method is based on linear genetic programming with the use of redundancy-removed recombination. The evolutionary synthesis can automatically construct feature extraction programs for a given object recognition problem without any domain-specific knowledge. An experiment was done on a lawn weed detection problem. The result shows that performances of the synthesized feature extraction program are comparable with those of the conventional lawn weed detection methods.

著者

松本 哲也 名古屋大学大学院情報科学研究科メディア科学専攻
竹内 義則 名古屋大学情報連携統括本部情報戦略室
工藤 博章 名古屋大学大学院情報科学研究科メディア科学専攻
大西 昇 名古屋大学大学院情報科学研究科メディア科学専攻
Matsumoto Tetsuya Department of Cardiovascular and Respiratory Medicine, Shiga University of Medical Science
WATCHAREERUETAI Ukrit Department of Media Science, Graduate School of Information Science, Nagoya University
TAKEUCHI Yoshinori Department of Media Science, Graduate School of Information Science, Nagoya University
KUDO Hiroaki Department of Media Science, Graduate School of Information Science, Nagoya University
OHNISHI Noboru Department of Media Science, Graduate School of Information Science, Nagoya University
Matsumoto Tetsuya Department Of Cardiovascular And Respiratory Medicine Shiga University Of Medical Science
大西 昇 名古屋大学 大学院 工学研究科 情報工学専攻
大西 昇 名古屋大学工学部
Ohnishi N Bio-mimetic Control Research Center Riken
Ohnishi Noboru Department Of Media Science Graduate School Of Information Science Nagoya University
Watchareeruetai Ukrit Department Of Media Science Graduate School Of Information Science Nagoya University
Matsumoto Tetsuya Department Of Cardiovascular And Respiratory Medicine
Matsumoto Tetsuya Department Of Informatics Kyushu University
Takeuchi Yoshinori Department Of Informatics And Mathematical Science Graduate School Of Engineering Science Osaka Univ
Ohnishi Noboru Department Of Information Engineering School Of Engineering Nagoya University
Kudo H Department Of Media Science Graduate School Of Information Science Nagoya University
Watchareeruetai Ukrit Dep. Of Media Sci. Graduate School Of Information Sci. Nagoya Univ.

関連論文

▼もっと見る