Low-cost multispectral camera for Precision Farming Application

Project Information

MCAPEFA

Grant agreement ID: 101028085

DOI

10.3030/101028085

Project closed

EC signature date 23 February 2021

Start date 1 September 2021

End date 31 August 2023

Funded under

• EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions

Total cost

€ 183 473,28

EU contribution

€ 183 473,28

Coordinated by

UNIVERSITA DEGLI STUDI DI TRENTO
Italy

Periodic Reporting for period 1 - MCAPEFA (Low-cost multispectral camera for Precision Farming Application)

Reporting period: 2021-09-01 to 2023-08-31

Summary of the context and overall objectives of the project

The MCAPEFA project addresses two issues. The first concerns designing a low-cost multispectral camera using only available components and open-source code. There is no need for specific tools or skills to assemble the developed device. Also, the proposed device is modular, and some components can be removed or changed for additional savings. This new camera contains nine narrow spectral bands in the visible and near-infrared range and one thermal infrared channel. It is 7 to 8 times cheaper than similar, commercially available devices. The second problem the MCAPEFA project addresses is detecting grapevine disease Flavescence dorée directly in the vineyard using hyperspectral and multispectral imaging. Flavescence dorée (FD) is one of the most important and damaging grapevine diseases that have severely threatened vineyards in recent years.
A multispectral camera is the most used device for collecting data for precision agriculture (PA), which is the key component of sustainable agriculture in the 21st century. Only with PA it is possible to produce more food and resources for other industries with less input while reducing ecological issues, especially environmental pollution, most prominent through the contamination of drinking water and aquatic ecosystems with unoptimized fertilization and pesticide usage. In the EU, most farms are small (86 % of all farms in the EU) and do not use precision agriculture to improve their productivity and reduce environmental pollution. The leading cause is a high initial investment in the equipment, especially cameras and drones. For example, in Italy, only 4.4 % of all small farms use drones in 2022. Therefore, it is crucial for global food security that small farms implement precision agriculture (PA) to become sustainable.
Furthermore, small farms are vital for preserving the territory and local rural areas' economic development. They guarantee the permanence of populations and agricultural activities in rural areas, contributing to incomes, social capital, local knowledge, and cultural heritage. Also, only PA can deal with climate change, especially drought and heat waves during spring and summer, when prolonged exposure to extremely high temperatures can negatively affect the plants. The MCAPEFA deals with this issue by providing the design of low-cost multispectral camera and reducing initial investments.
FD is an incurable, severe epidemic disease of grapevine in Europe caused by FD-phytoplasma. It is a quarantine disease with the obligation to uproot each infected plant to eliminate sources of FD. When the infection exceeds 20 % of all plants, the whole vineyard must be removed. Currently, the only available solution is to scout vineyards for infected plants. Trained agronomists and experts conduct it, but it is time-consuming, and each vineyard is controlled once in one or two years due to a lack of skilled individuals. And Flavescence dorée is constantly spreading around. For example, in Piedmont (Italy), from 2003 to 2018, FD increased from a very restricted zone to almost 25 % of the whole Piedmont grapevine area. In Trentino Alto-Adige (Italy), FD's presence nearly doubled in just one year (from 2021 to 2022).
The main goal of the MCAPEFA projects is to improve the adoption of precision farming in small farms by designing a low-cost multispectral camera and its use for detecting grapevine disease Flavescence dorée.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

The project started with research about the design of multispectral cameras described in the scientific literature, focusing on solutions that use multi-passband optical filters. Only such two realizations were published, but with approaches that resulted in a bulkier design than the initial MCAPEFA's idea. Next, the selection of components required for constructing a multispectral camera with an infrared thermal band was made, followed by the design of 3D printing. This resulted in an imaging device with nine narrow spectral bands in visible/near-infrared region and one thermal band. After assembling the proposed devices, software for controlling and image acquisition was prepared, followed by a calibration procedure and code for multispectral response estimation and image registration.
The second part started with data acquisition in the two vineyards near Riva del Garda (Italy), which was conducted three times during the summer of 2022, and one additional survey in the third vineyard during August 2023. The hyperspectral part of the data was used for testing the potential detection of Flavescence dorée (FD), a grapevine disease. Several tests showed that it is possible to differentiate hyperspectral profiles from healthy and infected leaves, even when the symptoms were slightly expressed and difficult to notice by experts. Additional acquisition showed that the proposed camera could be mounted on an unmanned aerial vehicle (UAV), and with obtained data, ortho-rectified images were successfully produced. Testing on the classification plant in healthy and infected by FD using acquired multispectral images is still under investigation.
The results of the MCAPEFA project were presented at 4 scientific conferences, which resulted in 3 peer-reviewed publications, 2 in conference proceedings, and one as journal articles. At Researchers' Night in Pavia (Italy), the findings from the MCAPEFA project were shown to a wider audience, and 8 presentations were organized for the students in Trento (Italy) and Belgrade (Serbia). During this research, two fairs were visited, and more than 25 contacts were established with the representatives of several companies interested in the use of designed multispectral camera or FD detection. Also, some results of the MCAPEFA project were shown at - MCAA Annual Conference 2023, while the project's progress was regularly reported using LinkedIn social media.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

The MCAPEFA project with the designed multispectral camera tries to increase the adaption of PA among small farms. Additionaly, this device can be used in many other fields, like waste sorting, food inspection, pharmaceutical products assessment, but also in scientific research as a low-cost tool for increasing knowledge about using multispectral imaging in various domains.
Next, all procedures for FD detection described in the literature are still in the research phase, and the proposed method differs in the way of data acquisition, presenting one more suitable to the field condition. Additionally, this approach can be adapted and implemented for scouting vineyards using standard equipment for field work, like tractors, resulting in a more economical way than using UAV (Unmanned Aerial Vehicle).
Additionally, as a multispectral camera is a basic tool for precision agriculture, this work contributes toward the aims of the strategy "Farm to fork" and to "A New Paradigm for Plant Nutrition": reduce by 50% the use and risk of chemical pesticides by 2030, reduce nutrient losses by at least 50%, while ensuring no deterioration on soil fertility, reduce fertilizer use by at least 20% by 2030. It should be highlighted that if the procedure for FD detection shows significant results in practice, especially for spreading FD and localizing infected zones, it will allow localized and direct treatment of contaminated vines, reducing soil and water pollution compared with the current procedure, which requires up to 10 spraying treatments yearly in the whole field.

The image shows the designed camera, logo, infected plants, data acquisition in the field and with