CC BY
27INTERNATIONAL SCIENTIFIC-PRACTICAL CONFERENCE ACTUAL ISSUES OF AGRICULTURAL DEVELOPMENT: PROBLEMS AND
SOLUTIONS
JUNE 6-7, 2023
REMOTE SENSING-BASED DATA ACQUISITION AND INFORMATION DRIVEN
AGRICULTURAL APPLICATIONS Kornél Szalay 1, Gabor Bércesi 2, Jozsef Deakvari 3, Jiri Soucek 4, Attila Lagymanyosi 5, Zoltan Bartfai6, Laszlo Katai 7, Istvan Szabo 8
1,2,3,5,6,7,8 Hungarian University of Agriculture and Life Sciences, Institute of Technology, Hungary, 2100 Godollo, Pâter Kâroly road 1.
4 Research Institute of Agricultural Engineering, p.r.i., Czeh-Republic, Drnovskâ 507 161
01 Prague 6
2bercesi.gabor@uni-mate.hu. +36/70-453-4792 https://doi.org/10.5281/zenodo.7996931
Abstract. Remote sensing is known as the most time- and cost-effective way of data collection. It plays decisive role in development of information based precision farming and environment protection. UAV or drone application is the latest technological implementation which integrates the capability of remote sensing and sight specific agricultural practice. Drone technology has high potential in many agricultural applications to save time and costs, to protect the environment and improve safety and ergonomics of sampling procedures, spraying and spreading.
Keywords: remote sensing, data acquisition, drones
Introduction The eye, the organ of vision developed during evolution to detect light, it performs remote sensing in the visible (380-750 [nm]) range of electromagnetic radiation. The visible range itself enables many analysis and classification procedures (Felfoldi et al., 2013). However, multi- and hyperspectral remote sensing devices make it possible to extend the visible range, thus displaying phenomena or information that are not visible to the human eye. Technology has created new horizons in the study of the environment.
Remote sensing The finding that forms the basics of the method is attributed to a Russian mineralogist and meteorite researcher (Krinov, 1947). Remote sensing is considered as a scientific activity in which the electromagnetic radiation reflected from the examined object or geographical area is measured from various distances using sophisticated sensors. The measured signal is converted into valuable information using various mathematical and statistical procedures. In the fast-paced world of modern civilization, remote sensing has become an essential tool for examining the balance and functioning of various natural and artificial systems. The basic condition of remote sensing is an energy source that illuminates the examined object or emitted from the object. Regardless of whether the energy source is natural (sunlight) or artificial (laboratory lighting) or emitted by the object itself we are talking about electromagnetic radiation. Talking about a point source of light - due to the Earth-Sun distance, the Sun is also a point source of light - the intensity of illumination is directly proportional to the power of the light source and inversely proportional to the square of the distance from the light source and it also depends on the angle of incidence. Remote sensing provides a number of methods and procedures for analysing various global and local processes. The huge amount of data is essential for studying global or local systems. Applications cover numerous agricultural, forestry, mining, urban and landscape planning, environmental protection, ecological, geological and hydrological applications, and is of particular importance in meteorological and climate change studies, as
INTERNATIONAL SCIENTIFIC-PRACTICAL CONFERENCE ACTUAL ISSUES OF AGRICULTURAL DEVELOPMENT: PROBLEMS AND
SOLUTIONS
JUNE 6-7, 2023
well as in military use. Remote sensing is based on the study of the interaction of material and light. During remote sensing, we can analyse a specific surface area, object, or event by collecting information without physical contact, within the original environment by avoiding any destruction or interventions (Lillesand et. al., 2004). In the beginning of 2000's the equipment used for measurements is divided into three categories according to their location. We distinguish systems used on the ground, in space and at different heights in the atmosphere. Sensor operating heights were defined are followings: Ground level (1-8 [m]), hang glider (100300 [m]), low height aircraft (300 [m] - 3 [km]), high height airplane (3-10 [km]), satellite (60035786 [km]). The following figure shows an illustration of remote sensing applications as of the year of 2000 (1. Figure) (Yamazaki, 2000).
1. Figure: Remote sensing applications The role of Unmanned Aerial Vehicles Together with the miniaturization of instrumentation and data systems the rapid development of UAVs as a remote sensing platform has resulted in an increasing uptake of this technology in the environmental and remote sensing science community. Regulations across the globe still limit the broader use of UAVs (Busznyâk, 2022). Their use is precision agriculture, ecology, atmospheric research, disaster response bio-security, ecological and reef monitoring, forestry, fire monitoring, quick response measurements for emergency disaster, Earth science research, volcanic gas sampling, monitoring of gas pipelines, mining plumes, humanitarian observations and biological/chemosensing tasks, continues to increase (2. Figure) (Tsourdos, 2017).
r
> 10,000 km
-1000 km
flj T3 -100 km
3
< 00 -10 km
G
^
o O. -1000 m
O (~1 km)
-100 m
< 1 m
Geostationary satellite
Remote sensing satellite
Fixed wing
Stratospheric platform
. Conv.
Rotary wing ^PP* Unmanned airship
Small UASs —
Ground-based
< 1 km2 10 km2 100 km2 1000 km2
Ground Coverage Area
> 1000 km2
2. Figure: A new horizon of remote sensing platforms
Unmanned Aerial Vehicle - UAV popularly known as drone, is an airborne system or an aircraft operated remotely by a human operator or autonomously by an onboard computer. There
INTERNATIONAL SCIENTIFIC-PRACTICAL CONFERENCE ACTUAL ISSUES OF AGRICULTURAL DEVELOPMENT: PROBLEMS AND
SOLUTIONS
JUNE 6-7, 2023
are two broad classes of UAVs - Fixed wing and Rotary based. When a UAV is equipped with sensors with remote sensing capabilities, it has the flexibility to gather information of various targets. Sensors for drones are increasingly being used for surveying, mapping, and inspections in several industries such as mining, construction, agriculture, environmental management, and waste management (NESAC, 2023). Fixed wing drones fly significantly longer, map larger areas, spend less time on-site, reduce labour costs, increase project capacity, simplify flight planning and operations and provide more flexibility with modular payloads. The major advantage of rotary-wing UAVs over fixed-wing UAVs is vertical take-off and landing. Rotary-wing UAVs are able to hover and change direction quickly. Lower speed, shorter range, and higher power consumption are considered as disadvantages. Hyperspectral remote sensing has been an important technical means to obtain detailed information for the quantitative analysis of environmental processes. Hypersepctral images have long been bound to complex flight campaigns performed by conventional airframes or to relatively lower resolution and the limited repeat coverage of multispectral satellites. Through the introduction of UAVs the collection of hyperspectral information becomes available for a wider community (Boxiong et al., 2022). Finally, various UAV-based applications were introduced into the agricultural practice. Hyperspectral remote sensing and analysis of agricultural areas are considered as the most developed technique in data collection to support the information-driven agricultural practice. UAV hyperspectral imagery has advantages over colour photography and multispectral remote sensing, with higher level of spectral details gives more sensitivity to physical-chemical properties. Snapshot hyperspectral imaging is a method to capture hyperspectral images during a very short integration time of a detector array. No scanning is involved and detectors consist of high number of pixels. Snapshot devices in general offer larger light collection capacity. Another great advantage of snapshot systems is the simplicity of imaging. Snapshot cameras are able to create the recording in a short time, which is why they can create images with a better signal-to-noise ratio. Applications have proved the potential of the technique in soil spectroscopy (Jung et al., 2015) and vegetation sciences. The most developed snapshot imaging hyperspectral camera to date is probably the Ultris X20 which is a 20 Megapixel spectral video camera, is sensitive in the UV, visible and near infrared range. With a wavelength range of 350-1000 nm, the Ultris X20 is the first UV-VIS-NIR hyperspectral video camera. Snapshot imaging spectrometer that produces 3D data cubes in real-time. This technology provides hyperspectral images with a spatial pixel resolution of 410*410, resulting in 168,000 pixels per frame and 164 spectral bands. Weighing less than 350 g, the camera is perfect for applications on small UAVs (Cubert, 2023).
Specifications ULTRIS X20
Wavelength Range 350 -1000 nm
Spectral Bands 164
FWHM Constant 10 nm
Spatial Resolution 410 x 410 pixel
Frame Rate 8 Hz
Weight 350g
4. Figure: Ultrix 20 specs.
Spraying and granular spreading
INTERNATIONAL SCIENTIFIC-PRACTICAL CONFERENCE ACTUAL ISSUES OF AGRICULTURAL DEVELOPMENT: PROBLEMS AND
SOLUTIONS
JUNE 6-7, 2023
There has been a great progress in vegetation mapping with various technologies (RGB, CIR, multi- and hyperspectral and thermal images). The imaging technology is available to map agricultural fields. UAV platforms can support a fast and accurate field survey (5. Figure) to identify various plant stresses originating of pests, diseases, drought or nutrient deficit. Emerging weeds spots can also be detected. The heterogeneity of soil can be mapped. Following the survey UAV has gathered the relevant information to define a treatment strategy of a particular field sprayer or spreader drones (WohnderDrone, 2023; DJI, 2023) can perform pest and weed control, seeding and fertilizer broadcasting missions (6. Figure, 7. Figure). The Matrice 300 RTK offers up to 55 minutes of flight time and advanced positioning. The multiple payload configurations provide simultaneously up to 3 payload mountings with a maximum capacity of 2.7 kg (DJI, 2023). The WohnderJet Agro H20 spraying system has been optimized for Drone Volt Hercules 20 industrial drone. It is possible to perform 6 hectares per hour spraying with an 8 litre per hectare dose (WohnderDrone, 2023). The AGRAS T40 has a spray load of 40 kg and a spread load of 50 kg (volume - 70 l). It supports multiple missions from surveying, mapping, to spraying and spreading, depending on the configuration used. It promises a 21.3 hectare per hour field spraying and 1.5 tonnes of fertilizer spreading capacity per hour (DJI, 2023). Airborne applications have a significant tradition in Hungary and other countries. Experience has been gathered over millions of hectares over decades. Conventional airborne agriculture operation has been labelled as obsolete technique facing challenges to meet the latest demands on accuracy and precision. A promising development has proved that retrofitting existing airframes with precision technologies can make a great difference: high-speed aerial applications can successfully transfer elements of precision farming to the air (Axial Ltd. and Forgoszarny Ltd., 2022). The Mi-2 helicopter (8. Figure) with unique developments made by Forgoszarny Ltd. provides a capacity to spread 9-11 tons of granules per hour or spray 120 ha in one hour (50l/ha dose) which is one magnitude higher than existing UAVs are capable of. The helicopter frame offers enough space to potentially operate a remote sensing system which could perform an on-flight analysis of the field to support a variable spraying/spreading dose rate. Such a solution would create a new horizon to perform mapping and spraying/spreading in one overflight.
INTERNATIONAL SCIENTIFIC-PRACTICAL CONFERENCE ACTUAL ISSUES OF AGRICULTURAL DEVELOPMENT: PROBLEMS AND
SOLUTIONS
_JUNE 6-7, 2023_
7. Figure: AGRAS T40 8. Figure: Precision airborne plant
protection (Mi-2)
Conclusions
Remote sensing is known as the time- and cost-effective way of data collection. It has decisive role in development of information based precision farming. Precision farming is an instrument to mitigate global challenges. UAVs have been rapidly developing in the last decade integrating the capability of remote sensing and sight specific agricultural practice. Today these techniques are more affordable and easy-to-use tools to improve the sustainability of plant production. Various sensors (RGB, multi- and hyperspectral cameras, thermal or LIDAR) can improve data collection and so the information-driven agricultural practice to increase yield and quality while minimizing environmental pollution, saving time and costs and improve safety and ergonomics of sampling procedures. The technical development of UAVs aims to increase payloads, flight time and to further improve spraying and spreading capacity. The continuous development of sensors and drones has greatly increased the amount of information which means more sophisticated data analysis methods are needed. There are visions of a future agricultural sector where UAVs have completely replaced field sprayers and conventional airframes, however, authors are proposing to take a complementary approach where available means of techniques are selected based on the local conditions and needs. UAVs have made available a wide range of remote sensing applications and should be considered as complementary tools to play their important role as a part of a complex system of tools in agricultural practice.
Acknowledgement The article was created with the support of the Thematic Excellence Program 2020 project TKP2020-NKA-16 and the long time development project of Research Institute of Agricultural Engineering p.r.i.
REFERENCES
1. Axial Kft., Forgoszarny Kft. (2022): Egyedulallo fejleszteseket jelentett be az AXIAL Kft. 2022. junius 3., /AXIAL Kft., Sajtokozlemeny https://www.agroinform.hu/gepeszet/egyedulallo-fejleszteseket-jelentett-be-az-axial-kft-56829-001
2. Boxiong Y. , Shunmin W., Shelei L., Bo Z., Fujun Z., Faizan A., Hui H. (2022):Research and application of UAV-based hyperspectral remote sensing for smart city construction. Cognitive Robotics Volume 2, 2022, Pages 255-266 https://www.sciencedirect.com/science/article/pii/S2667241322000234
3. Busznyak J. (2022a): Pilota nelkuli legijarmu-muveletek szabalyozasa. Mezogazdasagi Technika, 2022. februar, pp. 18-20
4. Cubert (2022): https://www.cubert-hyperspectral.com/
5. DJI (2023): https://www.dji.com/hu/
6. Felfoldi J., Baranyai L., Firtha F., Friedrich L., Balla C. (2013): Image processing based method for characterization of the fat/meat ratio and fat distribution of pork and beef samples. Progress in Agricultural Engineering Sciences Volume 9, Issue 1, 1 December pp. 27-53.
INTERNATIONAL SCIENTIFIC-PRACTICAL CONFERENCE ACTUAL ISSUES OF AGRICULTURAL DEVELOPMENT: PROBLEMS AND
SOLUTIONS
JUNE 6-7, 2023
7. Gili P., Civera M., Roy R., Surace C. (2021): An Unmanned Lighter-Than-Air Platform for Large Scale Land Monitoring. Remote Sens. 2021, 13(13), 2523; https://doi.org/10.3390/rs13132523
8. Jung, A., Vohland, M. and Thiele-Bruhn, S. (2015): Use of a Portable Camera for Proximal Soil Sensing with Hyperspectral Image Data. Remote Sensing, 7(9): 11434-11448 (2015).
9. Krinov E. L. (1947): Spectral reflectance properties of natural farmations. Laboratoriia aerometodov, Akad. Nauk SSSR. Moscow 1947. p. 271. In: National Research of Canada Technical Translation TT-439, http://nparc.cisti-icist.nrc-cnrc.gc.ca/npsi/ctrl?action=dsere&index=au&req="Krinov,+E.+L."
10. Lillesand T. M., Kiefer R. M., Chipman J. W. (2004): Remote sensing and image interpretation. 1.p. New York: Wiley, p. 763.
11. North Eastern Space Applications Centre (2023): UAV Applications. https://nesac.gov.in/scientific-programmes/remote-sensing-and-gis/uav-applications
12. Tsourdos A. (2017): Special Issue "UAV or Drones for Remote Sensing Applications". A special issue of Sensors (ISSN 1424-8220). https://www.mdpi.com/journal/sensors/special_issues/UAV_drones_remote_sensing
13. WohnderDrone Kft. (2023): https://www.wohnderdrone.hu/wohnderjet-agro-h20/
14. Yamazaki F. (2000): Applications of remote sensing and GIS for damage assessment. https://www.researchgate.net/figure/Platforms-and-sensors-of-satellite-and-airborne-remote-sensing
