Definition of damaged crops using satellite imagery in precision agriculture

Abstract

The relevance of identifying damaged crops using satellite imagery in precision agriculture is extremely high, as modern agriculture faces challenges such as climate change, pests, and natural disasters. The use of satellite data allows for rapid and accurate detection of problem areas in fields, contributing to more efficient resource management. Vegetation indices such as NDVI (Normalized Difference Vegetation Index) make it possible to identify the extent of crop damage, assess their condition, and determine the need for immediate intervention. This helps farmers reduce crop losses, lower field treatment costs, and simultaneously maintain ecological balance. Remote sensing technologies provide an environmentally friendly and rapid method for obtaining large volumes of data on the condition of agricultural land. Within the framework of precision agriculture, such methods become an integral part of the automation and modernization of the industry. Additionally, satellite imagery ensures continuous monitoring, which is particularly important for large-scale farms. The integration of this data into geographic information systems enables forecasting and optimization of agronomic processes. Thus, the use of satellite imagery is becoming a strategically important tool for enhancing the resilience of agriculture to modern challenges. The study aimed to determine the NDVI index for a cornfield and analyze which part of the crops had been damaged using medium spatial resolution satellite imagery of agricultural fields in the Lviv region. The research employed methods of photogrammetry and geoinformation analysis of Earth remote sensing data. To identify damaged crops in precision agriculture, medium spatial resolution images from the Landsat satellite were used for agricultural areas in the Lviv region. The analysis process included calculating the NDVI, which allows assessing the condition of crops based on their spectral characteristics. Input data were processed in the cloud environment of Google Earth Engine, where cloud masking, quality-based filtering, and NDVI vegetation index calculations for each pixel of the images were conducted. The collected data facilitated the creation of a damage map for the study area, simplifying the visualization and interpretation of results. The application of NDVI demonstrated the effectiveness of the method for precise monitoring of agricultural fields. During the period from spring to autumn 2023, changes in NDVI values were observed, which allowed identifying areas with signs of plant stress or damage. The highest index values indicated healthy vegetation, while lower values signaled damage or insufficient crop growth. The results showed that the technology enables agronomists to identify damaged areas and take timely measures to reduce losses and improve yield. The findings on identifying damaged crops using satellite imagery in precision agriculture can be used to optimize resource utilization, reduce crop losses, and improve the efficiency of agricultural production