Multilayer Perceptron Neural Network Approach to Estimate Chlorophyll Concentration Index of Lettuce (Lactuca sativa L.)

ODABAŞ M. S., Simsek H., Lee C. W., İşeri İ.

COMMUNICATIONS IN SOIL SCIENCE AND PLANT ANALYSIS, cilt.48, sa.2, ss.162-169, 2017 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 48 Sayı: 2
  • Basım Tarihi: 2017
  • Doi Numarası: 10.1080/00103624.2016.1253726
  • Dergi Adı: COMMUNICATIONS IN SOIL SCIENCE AND PLANT ANALYSIS
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.162-169
  • Anahtar Kelimeler: Artificial neural network, chlorophyll, leaf, lettuce, modeling, SPAD, REFLECTANCE, LEAVES, YIELD, IDENTIFICATION, ALGORITHM, QUALITY
  • Ondokuz Mayıs Üniversitesi Adresli: Evet

Özet

Nitrogen is an essential nutrient for greenhouse-grown lettuce (Lactuca sativa L.); however, excessive nutrient availability causes disease and detrimental effects on the leaf and root development. In this study, nitrogen content of the lettuce leaves was estimated by determining the chlorophyll concentrations of the leaves using image processing technique. The Hoagland solution was used as a fertilizer in five different doses (control, quarter of the solution, half of the solution, standard solution, and two times more of the solution). Multilayer perceptron neural network (MLPNN) model was developed based on the red, green, and blue components of the color image captured to estimate chlorophyll content and chlorophyll concentration index (SPAD values). According to the obtained results, the MLPNN model was capable of estimating the lettuce leaf chlorophyll content with a reasonable accuracy. The coefficient of determination was 0.98, and mean square error was 0.006 in validation process.