Determining the Growth Stage of Cut Gentian Flowers Using Ultraviolet Images

• Shono Hiroshi

  Iwate University

• Seki Asami

  Jouhougiken

• Yamaguchi Kyoko

  Tokyo Fresh

• Matsushima Uzuki

  Iwate University

• Koide Shoji

  Iwate University

• Takeda Jun-ichi

  Iwate University

Bibliographic Information

Other Title

• 切り花リンドウの生育ステージ判定における紫外線画像の有効性に関する検討

Abstract

Gentian (Gentiana spp.) is one of the most important floriculture crops grown in Iwate Prefecture, and production has steadily increased. They are generally used as cut flowers, but the flowers gradually deteriorate after being harvested. Because the commercial value of the flowers decreases as they deteriorate, discrimination of the growth stage can be an effective method to select flowers that will stay fresh as long as possible after being harvested. To date, however, no technique has been developed to discriminate the growth stage of gentian flowers. By examining the corollas of gentian flowers, we found that the spectral radiance of older samples was lower than that of younger samples in wavelengths between 700 and 900 nm. In this paper, we attempt to distinguish the growth stage of gentian flowers by measuring the pixel values of their corollas in ultraviolet images. The pixel values in ultraviolet images increased remarkably when the inner pollen was mature and had just begun to disperse. The pixel values decreased after this stage, possibly because the corollas were damaged by insects attracted to the pollen. We were thus able to use ultraviolet images of gentian flowers to determine the growth stage of gentian flowers and identify the best harvest time, which is before the plants are damaged by insects attracted to the pollen. <br>

Journal

• Agricultural Information Research

  Agricultural Information Research 18 (3), 122-129, 2009

  Japanese Society of Agricultural Informatics

Keywords

• <i>Gentiana</i>
• Ultraviolet Image
• Image Measurement
• Precision Farming

Details

• CRID: 1390282679436830848
• NII Article ID: 130000117374
• DOI: 10.3173/air.18.122
• ISSN: 18815219
• ISSN: 09169482
• Text Lang: ja
• Data Source:
  • JaLC
  • Crossref
  • CiNii Articles
  • KAKEN
• Abstract License Flag: Disallowed