Using Image Processing for Nitrogen Content Detection in Tomato Leaves

Kisalaei, Asma and Rasooli Sharabiani, Vali and Golmohammadi, Abdollah and Abbaspour-Gilandeh, Yousef and Esmailpour, Behrooz (2016) Using Image Processing for Nitrogen Content Detection in Tomato Leaves. Masters thesis, University of Mohaghegh Ardabili.

Text (کاربرد پردازش تصویر برای تشخیص میزان نیتروژن برگ گیاه گوجه فرنگی) Asma Kisalaei.pdf Download (655kB)

Abstract

Application of variable rate technology (VRT) of fertilizer on the farm is a major pillar of production accurate management that increases fertilization efficiency and reduces environmental pollution. The main purpose of this study is to investigate the possibility of determining of tomato nitrogen deficiency during the growing season using a digital camera and digital image processing. Using these methods causes less damage to the plant. Though, using this technology requires accurate and continuous determination of plant nitrogen status on the farm during of growth stages. Therefore 18 plots of Spring tomato by six different levels of fertilization (0, 60, 120, 180, 240, 300 Kg/ha) with three repetition in completely randomized designs were sown in plastic pots with a diameter of 25 cm. It was attempted to be fertilized after the growing wheat in generative step of plant. The nitrogen content of leaves was measured by two methods; 1-using chlorophyll meter (SPAD) and 2- doing Kjeldahl test, in two stages, once before fertilization and the other time one week after fertilization. In addition, with cutting the samples of plant leaves, it was attempted to take pictures using a digital camera in the chamber that was controlled in terms of light, simultaneously measuring the nitrogen content. All pictures were transferred to the space on MATLAB R2013 software and average color of R, G, B and 13 color models were extracted from RGB images. To evaluate the processed results of color parameters and models with two measuring methods; SPAD Index and Kjeldahl test, the statistical analysis (the correlations, variance analysis (ANOVA) and compared means) were considered. The results showed that the G-band have the best correlation with the results of chlorophyll content (SPAD) and nitrogen content (Kjeldahl test) in both stages of measurements. The highest coefficient of determination was related to the chlorophyll content (SPAD) and G-band in the second stage of measurement (R2=0.96). The same amount for G-band was correlated by present nitrogen content (R2=0.97).

Actions (login required)

View Item