Use of Drones in Precision Agriculture

Precision agriculture increasingly relies on Unmanned Aerial Vehicles (drones) to monitor crops and detect pests and diseases early. These aerial systems collect real-time data on plant health, allowing farmers to act quickly and accurately. For example, a farmer flies a field with a drone equipped with a high-resolution camera; the device takes detailed images of the crop and transmits the data for immediate analysis. With this information, it is possible to identify areas with water stress, nutritional deficiencies, or pathogen outbreaks before symptoms are visible to the human eye. In summary, drones offer a real-time, comprehensive view of the crop, optimizing input use and reducing losses due to pests.

Advanced Sensors in Agricultural Drones

Modern drones incorporate various remote sensors that detect information outside the visible spectrum. The most common include:

• RGB Camera (Visible Spectrum): Captures high-resolution color images for basic visual analysis, weed detection, and assessment of plant vigor.

• Multispectral Camera: Records near-infrared bands and other wavelengths invisible to the human eye. These are used to calculate vegetation indices (e.g., NDVI) that indicate photosynthetic activity. These indices alert to stress from pests, diseases, or nutrient deficiencies before they are apparent (1).

• Thermal Camera (Thermal Infrared): Measures plant and soil temperature. Sick or water-stressed plants radiate more heat, so thermal images can pinpoint pest or fungus outbreaks before the damage is visible (1).

• LiDAR: Scans the terrain in $3\text{D}$ with a laser, useful for understanding topography, modeling drainage, and planning irrigation or drainage systems. While not directly used for pest detection, it complements advanced agricultural mapping (1).

In addition to these, some drones may carry high-precision multispectral or hyperspectral sensors to capture more light bands, and integrated meteorological equipment (humidity, temperature). Thanks to this range of sensors, a drone can map large areas in a few hours and generate georeferenced orthomosaics of the field, which are essential for subsequent analysis (1, 2).

Image Analysis and Artificial Intelligence

Images collected by drones are processed using specialized software and Artificial Intelligence (AI). Computer vision and machine learning algorithms analyze patterns of color, texture, and shape to identify damage. For example, convolutional neural networks are trained with thousands of photos of healthy and diseased leaves; in this way, the system learns to recognize incipient symptoms. In a field study, a neural network prototype was built with multispectral images and achieved $85\%$ accuracy in detecting plant stress before visible wilting (1, 3).

Another typical analysis is the calculation of indices like the NDVI (Normalized Difference Vegetation Index). This index compares the infrared and visible reflectance of vegetation and allows for detecting pests or water deficits in a plot. Areas with a low NDVI usually correspond to weakened plants. Thus, NDVI is fundamental for locating problematic areas throughout the crop (4). AI models can also generate heat maps and automatic alerts that indicate disease hotspots. By combining multispectral, thermal, and other sensor data, AI systems provide precise diagnoses that help in making decisions (localized irrigation, specific phytosanitary treatment) only where necessary (3, 4).

Pest and Disease Monitoring

The use of drones allows for mapping the incidence of pests and diseases with millimetric precision. After each flight, image analysis generates maps that distinguish healthy plants from affected plants. For example, mildew spots or pests in a vineyard can be flagged long before the farmer notices symptoms. A real case demonstrated this: in a vineyard in Castilla-La Mancha, early detection with multispectral images located disease outbreaks and allowed for treating only $5\%$ of the surface, instead of the usual $100\%$ (5). This not only drastically reduced the use of pesticides but also preserved beneficial fauna in the crop by avoiding total fumigation (5).

In practice, drones regularly fly over the plantation (e.g., weekly), and each flight provides a new, updated image mosaic. Agronomists use these maps to adjust treatments: applying fungicides or insecticides only to the affected "patches," not to the entire field. This achieves localized control of diseases and pests, minimizing environmental and pharmacological impacts. Additionally, thermal cameras can quickly indicate areas with water stress associated with diseases. Altogether, these systems achieve a high-resolution “pest map” with each flight (6), which speeds up the taking of corrective measures.

Maps produced from multispectral and thermal data clearly visualize areas with plant stress or pathogen presence, thus facilitating localized interventions.

Economic Impact and Return on Investment

The introduction of drones in agriculture has significant economic benefits. These include:

• Increased Yield: Precision agriculture using drones can significantly increase crop productivity. For example, a case reported in Mexico achieved a $\mathbf{30\%}$ increase in banana production thanks to analysis with aerial data (7). In general, drone-monitored crops usually show higher yields because management decisions (irrigation, fertilization, phytosanitary control) are based on accurate data.

• Reduced Inputs: By applying fertilizers and pesticides only where needed, chemical expenditure is reduced. Studies indicate that the focused use of agrochemicals can cut pesticide and fertilizer consumption by up to $20-40\%$, by eliminating unnecessary applications (8, 9). The citation from the rice study mentioned that treating only the diseased areas "eliminates the current method... costs in purchasing chemical products will be reduced" (8).

• Water Savings: Irrigation optimization is another economic effect. Through humidity sensors and data analysis, excessive irrigation is avoided. For example, in an olive grove, a $25\%$ water saving was achieved thanks to smart irrigation, also improving oil quality by avoiding prolonged water stress (10). In general, localized irrigation reduces both the volume of water used and associated energy costs.

• Reduced Operating Costs: Drones replace manned flights or walking inspections, shortening inspection times. According to specialized sources, the cost of a drone flight is much lower than that of a traditional plane or helicopter (11). Furthermore, by detecting pests before they spread, significant harvest losses are avoided. All of this accelerates the Return on Investment (ROI) in drone technology. In fact, it has been reported that many farms recover the investment in the first year or even the same harvest after incorporating drones (11).

Collectively, these improvements translate into clear economic gains for the producer. For example, reducing pesticide use by $30\%$ implies a direct saving in inputs; meanwhile, a $15-30\%$ increase in yield increases revenue. Drone technology usually has a favorable ROI, especially in medium to large-scale farms where the benefits per hectare are amplified (11). Furthermore, being a reusable tool, the marginal costs of each additional flight are low, making the investment increasingly profitable over time.

Conclusions

In summary, drones equipped with advanced sensors and supported by artificial intelligence offer an unprecedented capacity to monitor pests and diseases in various crops. They provide multispectral, thermal, and RGB images that, processed with AI algorithms, allow for the detection of phytosanitary problems before they are evident. This enables the application of localized treatments, reducing agrochemical costs and improving sustainability. Numerous studies and real cases confirm that precision agriculture with drones increases yields and the return on investment. Although there is an initial investment in equipment and training, data analysis shows that the benefits in productivity, input savings, and efficient management make drones a key tool for the modern farmer (8, 11).

References: Technical data and cited cases based on recent studies and industry reports (1, 4, 5, 6, 7, 8, 11).

(1, 3, 5, 10) La Inteligencia Artificial en la agricultura: drones y sensores

https://academiaruraldigital.es/ia-en-agricultura-drones-sensores/

(2, 4, 6, 11) Drones en la agricultura,-Ag Tech-Tecnología Hortícola

https://www.tecnologiahorticola.com/drones-agricultura-tecnologia/

(7) Investigación y desarrollo para agricultura | Pix4D

https://www.pix4d.com/es/industria/agricultura/gestion-agricola

(8) (PDF) Agricultura de precisión con drones para control de enfermedades en la planta de arroz

https://www.researchgate.net/publication/348421662_Agricultura_de_precision_con_drones_para_control_de_enfermedades_en_la_planta_de_arroz

(9) ¡La Revolución de los Drones en la Agricultura! Un Análisis Completo

https://hablemosdedrones.com/hablemosdedrones-com-drones-agricultura-inteligente-2025/