ニューラルネットワークを用いた双眼視画像装置の改良 Improvement of a Binocular Stereovision System by Using Artificial Neural Networks

この論文にアクセスする

この論文をさがす

著者

抄録

A binocular stereovision system has been developed to estimate growth variables of a transplant population. In the present study, the image analysis system was improved by adopting a three-layered artificial neural network model (ANN model) based on a back-propagation algorithm. Inputs of the ANN model were average height, leaf area, projected leaf area, and mass volume of the transplant population obtained from the image analysis system. Outputs of the ANN model were average height, number of unfolded leaves, leaf area, and fresh and dry masses of the transplant population, which give a more accurate assessment of the transplant growth status than that obtained from the image analysis system. The number of nodes in the hidden layer of the ANN model was determined through trial and error. The growth variables thus obtained from the ANN model using a sweetpotato (Ipomoea batatas (L.) Lam.) transplant population were more accurate than those obtained from a regression model. The image analysis system, after being improved by using the ANN model, successfully identified the transplant growth status with a high degree of accuracy.

A binocular stereovision system has been developed to estimate growth variables of a transplant population. In the present study, the image analysis system was improved by adopting a three-layered artificial neural network model (ANN model) based on a back-propagation algorithm. Inputs of the ANN model were average height, leaf area, projected leaf area, and mass volume of the transplant population obtained from the image analysis system. Outputs of the ANN model were average height, number of unfolded leaves, leaf area, and fresh and dry masses of the transplant population, which give a more accurate assessment of the transplant growth status than that obtained from the image analysis system. The number of nodes in the hidden layer of the ANN model was determined through trial and error. The growth variables thus obtained from the ANN model using a sweetpotato (<I>Ipomoea batatas</I> (L.) Lam.) transplant population were more accurate than those obtained from a regression model. The image analysis system, after being improved by using the ANN model, successfully identified the transplant growth status with a high degree of accuracy.

収録刊行物

  • 植物工場学会誌

    植物工場学会誌 14(1), 18-24, 2002-03-01

    日本植物工場学会

参考文献:  13件中 1-13件 を表示

各種コード

  • NII論文ID(NAID)
    10012339059
  • NII書誌ID(NCID)
    AN10451761
  • 本文言語コード
    ENG
  • 資料種別
    ART
  • ISSN
    09186638
  • NDL 記事登録ID
    6087058
  • NDL 雑誌分類
    ZR3(科学技術--生物学--植物)
  • NDL 請求記号
    Z18-3397
  • データ提供元
    CJP書誌  NDL  IR  J-STAGE 
ページトップへ