Improved food security requires a deeper knowledge of the mechanics of food production. Increasing agricultural yields has been shown to have a substantial impact on alleviating poverty. Several variables affect yield, the amount of harvested agricultural product in a given area. Technological (farming methods, management choices, etc.), biological (diseases, insects, pests, weeds), and environmental variables are the standard classifications for these elements (climatic condition, soil fertility, topography, water quality, etc.) . Differences in yield in regions throughout the globe may be traced back to these causes.

Knowing how to predict grain output and crop loss early on and accurately is a crucial ability in the grain industry. Estimates need to be precise in order to help farmers with:

• for the sake of crop insurance • preparing ahead for harvesting and storing • forecasting the cash flow

Early-stage yield estimation requires a great deal of hands-on expertise. Accurate yield predictions may be made as crops approach maturity.

Data gathered by satellites for use as remote sensing (RS) has been used for a variety of scientific inquiries and practical purposes. Meteorology, canopy and soil research, agriculture and crop production, water, ice, and ocean research and management, geology, mapping, land use and environmental monitoring, reconnaissance, defense, etc. are among the most significant areas of use.

The primary objective of this study was to create a technique for yield estimate that relies (during the active phase) only on RS data collected by satellite. As a consequence of our prior work and observations, we came up with this concept . There are often two distinct times in every yield prediction procedure. Procedure calibration comes first, followed by actual implementation. The first stage makes use of ground-truth data, but the second stage, when the system is actually put into service, makes no use of such information. We calibrate using ‘historical’ field-level agricultural outputs from counties, regions, farms, and in this case, the Central Statistical Office of Hungary (KSH). Adding the actual year’s data to the database (often at the conclusion of the agricultural year) and fine-tuning the model if required are both parts of the calibration process, which overlaps with the operational phase

In conclusion, the robust technique proven to be a useful tool in regional yield estimate, and its reliability and accuracy are generally acceptable. After a recalibration procedure, such as determining the crop-filters for the provided regions and the relief-stratum correction terms, the robust approach may be used in places outside Hungary. These strategies are flexible enough to be used in both developed and developing nations due to their low cost and reliance on low-tech solutions.