RESEACH ON PRESUMPTION OF THE SULFIDE CONDENTRATION OF THE WATERS BOTTOM MUD SURFACE BY A BOTTOM MUD COLOR SYSTEM
| Autor: | Koichi, Yakita, Kiyoshi, Takikawa, Kazumi, Akimoto, Tatsuya, Masuda, Kentaro, Morimoto, Keisuke, Mori, Hideyuki, Shimasaki |
|---|---|
| Jazyk: | japonština |
| Rok vydání: | 2011 |
| Předmět: | |
| Zdroj: | 熊本大学工学部技術部年次報告集. 2011:141-146 |
| Popis: | 八代海域で採集した試料を対象に、海域底泥表層の硫化物濃度を簡便・効率的に推定する手法について、土色計およびデジタルカメラ画像のマンセル、L*a*b*, HSV等の表色系の色彩データを用いて基礎的検討を行った。その結果、土色計で得られたマンセル色相、L*, a*, b*色彩データを説明変数とする重回帰分析とa*,b*色彩データを説明変数とした最良回帰式で有意性が認められ、硫化物濃度推定の可能性が示された。底泥表面に非接触で校正板を必要としない手法の検討では、デジタルカメラ画像の各色彩データを主成分分析し、説明変数としてB, L*, a*, b*, Hを選定した。これらを説明変数とした最良回帰式より、推定式の有意性は認められ、デジタル画像からの推定式の可能性も示された。また、色相a*が硫化物濃度の推定式に与える影響が大きいことが分かり、土色計の説明変数と共通する結果となった。 For the sample collected in Yashiro bay, the color data of color systems, such as Mansel of soil color reader and a digital camera picture, L*a*b*, and HSV, was used for the sulfide concentration on the bottom mud surface of submarine about the presumed technique simple and efficiently, and fundamental examination was performed. As a result, significance was accepted by the multiple linear regression analysis and the best regression which make an explaining variable the color data obtained by soil color reader. By the technique of not needing proofreading by non-contact, principal component analysis of the color data of a digital camera picture was conducted, and B, L*, a*, b*, and H were selected as an explaining variable. From the best regression which made these the explaining variable, the significance, of the estimate equation was accepted and also showed the possibility of the estimate equation from a digital image. Moreover, it turned out that the influence which a* has on the estimate equation of sulfide concentration is great, and resulted in it being common in the explaining variable of soil color reader. However, if F value is not large even if analysis of variance performs f official approval and significance is found out, it is not effective in actually using. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|