Ademir Ferreira da Silva¹, Ricardo Luis Ohta², Jaione Tirapu Azpiroz^1*, Matheus Esteves Fereira¹, Daniel Vitor Marçal¹, André Botelho², Tulio Coppola², Allysson Flavio Melo de Oliveira², Murilo Bettarello³, Lauren Schneider³, Rodrigo Vilaça⁴, Noorunisha Abdool⁵, Pedro Augusto Malanga⁶, Vanderlei Junior⁶, Wellington Furlaneti⁶, Mathias Steiner^1*

1 IBM Research - Avenida República do Chile, 330, Rio de Janeiro, RJ 20031-170, Brazil

2 IBM Research - Rua Tutoia 1157, Sao Paulo, SP 04007-900, Brazil

3 ENVERITAS Non-profit Organization, New York, NY, USA

4 CSEM BRASIL - Avenida José Candido da Silveira, 2000, Belo Horizonte, MG 31035-536, Brazil

5 OMNIA FERTILIZERS RSA, Bryanston, Sandton, South Africa

6 Integrada Agroindustrial Cooperative - R. São Jerônimo, 200 - Centro, Londrina, PR 86010-480, Brazil

* Corresponding authors emails: jaionet@br.ibm.com, mathiast@br.ibm.com

DOI10.24435/materialscloud:c6-39 [version v3]

Publication date: Jun 28, 2022

How to cite this record

Ademir Ferreira da Silva, Ricardo Luis Ohta, Jaione Tirapu Azpiroz, Matheus Esteves Fereira, Daniel Vitor Marçal, André Botelho, Tulio Coppola, Allysson Flavio Melo de Oliveira, Murilo Bettarello, Lauren Schneider, Rodrigo Vilaça, Noorunisha Abdool, Pedro Augusto Malanga, Vanderlei Junior, Wellington Furlaneti, Mathias Steiner, Artificial intelligence enables mobile soil analysis for sustainable agriculture, Materials Cloud Archive 2022.86 (2022), https://doi.org/10.24435/materialscloud:c6-39

Description

For optimizing production yield while limiting negative environmental impact, sustainable agriculture benefits greatly from real-time, on-the-spot analysis of soil at low cost. Colorimetric paper sensors are ideal candidates for cheap and rapid chemical spot testing. However, their field application requires previously unattained paper sensor reliability and automated readout and analysis by means of integrated mobile communication, artificial intelligence, and cloud computing technologies. Here, we report such a mobile chemical analysis system based on colorimetric paper sensors that operates under tropical field conditions. By mapping topsoil pH in a field with an area of 9 hectares, we have benchmarked the mobile system against precision agriculture standards following a protocol with reference analysis of compound soil samples. As compared with routine lab analysis, our mobile soil analysis system has correctly classified soil pH in 97.8% of cases (132 out of 135 tests) while reducing the analysis turnaround time from days to minutes. Moreover, by performing on-the-spot analyses of individual compound sub-samples in the field, we have achieved a 9-fold increase of spatial resolution that reveals pH-variations not detectable in compound mapping mode. Our mobile system can be extended to perform multi-parameter chemical tests of soil nutrients for applications in environmental monitoring at marginal manufacturing cost. This record comprises a data set of approximately 800 images of a colorimetric paper-based chemical analysis device captured with a custom mobile application at outdoor conditions. The data set also contains three csv files collecting the color information (RGB) values and analysis results (pH values) as determined by the mobile application models from those paper-based device images. These csv files correspond to the RGB data and associated pH values as captured on the field (pre-processed), after processing post field test and a set corresponding to the measurements on a compound sample combining the 9 soil samples collected per hectare. An additional data set is included corresponding to the images and RGB data used for the calibration of the logistic regression models used by the mobile application to predict pH values from the colorimetric information. Python code is made available for the analysis of the colorimetric chemical reaction on paper of chemical reagents to soil samples across a range of acidity values, including the application of an illumination compensation method, and for the training of predictive machine learning models.

Materials Cloud sections using this data

Files

File name	Size	Description
All Original Figures.zip MD5md5:312582fb9b9c5bcf8353ce0c9dea10d5	294.1 MiB	Data set of images of colorimetric paper-based chemical analysis device captured with a custom mobile application.
FieldTestAllData_external_byID_preProcessed.csv MD5md5:1f3eccb410e0552142d5cf5aa8a5141e	125.6 KiB	csv file collecting the color information (RGB) values and analysis results (pH values) as determined by the mobile application models. All data.
FieldTestAllData_external_byID_postProcessed.csv MD5md5:47aede01f60142e5f0f584f664396ad0	126.5 KiB	csv file collecting the color information (RGB) values and analysis results (pH values) as determined by the mobile application models. After processing
FieldTestAllData_external_byID_CompoundSamples.csv MD5md5:3b8178529683a65e3c4ab93e9524dafb	8.6 KiB	csv file collecting the color information (RGB) values and analysis results (pH values) as determined by the mobile application models. Compound samples per hectare.
CalibrationData.zip MD5md5:67c2db727d5f1c7670577506b9553d1d	42.4 KiB	csv files with the RGB values corresponding to colorimetric paper-based chemical analysis devices used for the calibration of the logistic regression models used by the mobile application to predict pH values from the colorimetric information.
CalibrationImages.zip MD5md5:a3c496ad0eb68f731cb95c6025e39abb	2.8 GiB	Dataset of images corresponding to colorimetric paper-based chemical analysis devices used for the calibration of the logistic regression models used by the mobile application to predict pH values from the colorimetric information.

License

Files and data are licensed under the terms of the following license: MIT License.
Metadata, except for email addresses, are licensed under the Creative Commons Attribution Share-Alike 4.0 International license.

External references

Keywords

colorimetric sensor paper-based device soil testing acidity field test mobile application Experimental