Erdem Emin MARAŞ, Mehmet Serhat ODABAS, Sreekala BAJWA, Chiwan LEE

The Prediction of Saint John's Wort Leaves' Chlorophyll Concentration Index using Image Processing with Artificial Neural Network

Bitki azot içeriğinin tespiti için bitkisel analizlerin dahil olduğu yaprak klorofil ölçümü ve uzaktan algılama tekniklerin de dahil olduğu çeşitli yöntemler mevcuttur. Bu çalışmada, görüntü işleme yöntemi ile kantaron (Hypericum perforatum L.) yapraklarının klorofil konsantrasyonu tahmin edilmiştir. Araştırmada, saksılarda yetiştirilen kantaronlara 5 farklı dozda Hougland solusyonu gübre olarak uygulanmıştır. Yaprakların klorofil konsantrasyonunun ölçülmesinde SPAD-502 klorofil metre kullanılmıştır. UV spektrometresi ile yaprakların klorofil-a (CHL-a) ve klorofil-b (CHL-b) içerikleri ölçülmüştür. Yapay Sinir Ağı (YSA) modeli kullanılarak klorofil konsantrasyonunu tahmin etmek için bir dijital kamera ile çekilen renkli görüntülerin RGB (kırmızı, yeşil ve mavi) bileşenlerinden faydalanılmıştır. Sonuç olarak yapay sinir ağı ile yüksek doğrulukta kantaron yapraklarının klorofil konsantrasyonunu tahmin edilmiştir. Doğrulama R2 0.99 ve MSE 0.005 olarak elde edilmiştir .Bu değerler yapay sinir ağı modelinin güvenirliliğini ortaya koymaktadır.

Yapay Sinir Ağı ile Görüntü İşleme Kullanarak Kantaronda Klorofil Konsantrasyon Endeksi Tahmini

There are several methods for detecting plant nitrogen content including plant analysis like leaf chlorophyll measurement and remote sensing techniques. In this study, image-processing method was used to predict St. John's wort (Hypericum perforatum L.) leaf chlorophyll concentration from leaves. The experiment was carried out in greenhouse conditions. The Hougland solution was used as a fertilizer. It was applied at 5 different levels to the St. John's wort grown in pots. SPAD-502 chlorophyll meter was used for measuring the chlorophyll concentration of the leaves. The chlorophyll-a (chl-a) and chlorophyllb (chl-b) of the leaves were measured by UV spectrometer. Artificial Neural Network (ANN) model was developed based on the RGB (red, green, and blue) components of the color image captured with a digital camera for estimating the chlorophyll concentration. According to the obtained results, the neural network model is capable of estimating the St. John's wort leaf chlorophyll concentration with a reasonable accuracy. The coefficient of determination (R2) was 0.99 and mean square error (MSE) was obtained 0.005 from validation.

PDF

___

Chappelle EW, Kim MS, McMurtrey JE (1992). Ratio analysis of reflectance spectra (RARS): an algorithm for the remote estimation of the concentrations of chlorophyll A, chlorophyll B, and carotenoids in soybean leaves. Remote Sensing of Environment. 39: 239-247.
Chen CT, Chen S, Hsieh KW, Yang HC, Hsiao S, Yang C (2007). Estimation of leaf nitrogen content using artificial neural network with cross-learning scheme and significant wavelengths. Transactions of the ASAE. 50: 295-301.
Cirak C, Ayan AK, Odabas MS (2011). Seed Germination of Hypericum triquetrifolium and Hypericum heterophyllum L. Medicinal and Aromatic Plant Science and Biotechnology. 5 (11): 105-107.
Demuth H, Beale M (2000). Neural network toolbox, Version 4., The MathWorks, Inc. Natick, MA, USA.
Gautam RK, Panigrahi S (2007). Leaf nitrogen determination of corn plant using aerial images and artificial neural networks. Canadian Biosystems Engineering. 49: 71-79.
Kawashima S, Nakatani M (1998). An algorithm for estimating chlorophyll content in leaves using a video camera. Annals of Botany. 81: 49-54.
Koumpouros Y, Mahaman BD, Maliappis M, Passam HC, Sideridis AB, Zorkadis V (2004). Image processing for distance diagnosis in pest management. Computers and Electronics in Agriculture. 44: 121-131.
Markwell J, Osterman JC, Mitchell JL (2005). Calibration of the minolta SPAD-502 leaf chlorophyll meter. Photosynthesis Research. 16: 467-472.
Namrata J, Ray SS, Singh JP, Panigrahy S (2007). Use of hyperspectral data to assess the effects of different nitrogen applications on a potato crop. Precision Agriculture. 8: 225-239.
Noh H, Zhang Q, Shin B, Han S, Feng L (2006). A neural network model of maize crop nitrogen stress assessment for a multi- spectral imaging sensor. Biosystems Engineering. 94: 477-485.
Odabas MS, Temizel KE, Caliskan O, Senyer N, Kayhan G, Ergun E (2014). Determination of reflectance values of hypericum's leaves under stress conditions using adaptive network based fuzzy inference system. Neural Network World. 24 (1): 79-87.
Odabas MS, Leelaruban N, Simsek H, Padmanabhan G (2014). Quantifying impact of droughts on barley yield in north dakota, usa using multiple linear regression and artificial neural network. Neural Network World. 24 (4): 343-355.
Pagola M, Ruben O, Ignacio I, Humberto B, Edurne B, Pedro AT, Carmen L, Berta L (2009). New method to assess barley nitrogen nutrition status based on image color analysis comparison with SPAD-502. Computers and Electronics in Agriculture. 65: 213- 218.
Pydipati R, Burks TF, Lee WS (2006). Identification of citrus disease using color texture features and discriminate analysis. Computers and Electronics in Agriculture. 52: 49-59.
Temizel KE, Odabas MS, Senyer N, Kayhan G, Bajwa S, Caliskan O, Ergun E (2014). Comparision of some models for estimation of reflectance of hypericum leaves under stress conditions. Central European Journal of Biology. 9(12): 1226-1234.
Wang Y, Wang F, Huang J, Wang X, Liu Z (2009). Validation of artificial neural network techniques in the estimation of nitrogen concentration in rape using canopy hyperspectral reflectance data. International Journal of Remote Sensing. 30: 4493-4505.
Wellburn AR (1994). The spectral determination of chlorophylls a and b, as well as total carotenoids, using various solvents with spectrophotometers of different resolution. Journal Plant Physiology. 144: 307-313.
Woebbecke DM, Meyer GE, Bargen KV, Mortensen DA (1995). Color indices for weed identification under various soil, residue and lighting conditions. Transactions of the ASAE. 38: 259-269.