تقدير محصول القمح في مزرعة محافظة النجف/العراق باستخدام أزواج صور الأقمار الصناعية متعددة الأزمنة

الملخص

The present work aims to estimate the yield of wheat crop for a particular wheat farm in Najaf Governorate, southwestern Iraq.
To achieve this objective, seven multispectral bands of the Landsat satellite with a spatial resolution of 30×30 square meters were used for two different time periods. The satellite image for the first time represents the high vegetative growth of the wheat crop, and the satellite image for the second time represents the farm after harvest.
The RGB coloring model was used as an unsupervised method, which classifies the scene according to its three-component RGB color combination. This model was used to determine the number of classes in a scene. Depending on the number of classes determined by the unsupervised method, a supervised classification using a maximum-likelihood technique was used to perform the classification process for the Multi temporal-spectral images. A mask was created to define the selected area (a wheat farm) and through it, the data was read within the boundaries of the area in which the yield will be calculated.
The results show that the rate of wheat production for the study area is equal to 89.27 ton by using spectral bands. It can also be seen that the average percentage error (8.66%) applied with the spectral bands.

Keywords: Wheat production, remote sensing, multi-temporal data, crop yield.