A unified exploratory data analysis system for potato breeding and monitoring using IoT and UAV data
Abstract
Breeding programs increasingly require high-resolution, continuous monitoring to evaluate potato varieties under rapidly changing climatic conditions. However, existing Exploratory Data Analysis (EDA) tools rarely integrate heterogeneous data streams – UAV imagery, IoT sensing, and field observations – into a unified, automated, and breeder-friendly workflow. This fragmentation limits breeders’ ability to detect stress, compare genotypes, and make timely, evidence-based decisions. We present a unified EDA system that consolidates multispectral UAV data, in-field IoT temperature measurements, historical agronomic records, and GIS-based spatial analyses into a single end-to-end architecture. The platform automates data acquisition, ETL processes, quality control, and relational storage while offering an interactive interface for flexible visualization and multi-scale analysis. The system was deployed over two growing seasons in Latvia using two potato varieties (Brasla and Imanta), 10 UAV missions per year, 4,800 multispectral images per season, and continuous five-minute sensor data. Field evaluations with professional breeders show that the system enhances data exploration, improves usability, and reduces information loss compared with traditional manual workflows. A qualitative assessment of the interface indicated greater user confidence and clearer access to key indicators. To address the absence of established benchmarks, we provide quantitative metrics demonstrating data quality, precision, automation performance, and integration reliability. Comparisons of our NDVI outputs with values reported in the literature further validate the robustness of the multispectral processing pipeline. Overall, this work introduces the first unified, breeder-oriented EDA framework that combines high-precision monitoring, automated multi-source integration, and interactive analytics. The results demonstrate how accessible digital tools can accelerate varietal selection and support more resilient and sustainable potato breeding under increasing environmental variability.
Related
- Development of a lightweight broadband cavity-enhanced spectrometer system for simultaneous monitoring of HONO and NO2
- Assessment of an updated polar stratospheric cloud parameterisation for the UK Earth System Model (UKESM1.1) within the UK Met Office Unified Model (v13.9) using CALIOP and MLS observations
- A coordinated NASA observing system for monitoring the next decade of ice sheet-climate interactions
- A low-cost surface ablation monitoring system using wind-induced motion of mass-balance stakes
- Long-term monitoring of a full-scale parallel-unit downflow hanging sponge system with adaptive capacity for fluctuating wastewater flows in Japan
BibTeX
@article{lapans2026unified,
title = {{A unified exploratory data analysis system for potato breeding and monitoring using IoT and UAV data}},
author = {Andris Lapans and Jean-Baptiste Pichancourt and Sandro Bimonte and Ginta Majore},
journal = {PeerJ Computer Science},
year = {2026},
month = jul,
doi = {10.7717/peerj-cs.4002},
url = {https://hal.science/hal-04684062v3},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - A unified exploratory data analysis system for potato breeding and monitoring using IoT and UAV data
AU - Andris Lapans
AU - Jean-Baptiste Pichancourt
AU - Sandro Bimonte
AU - Ginta Majore
PY - 2026
DA - 2026/07/16
JO - PeerJ Computer Science
DO - 10.7717/peerj-cs.4002
UR - https://hal.science/hal-04684062v3
AB - Breeding programs increasingly require high-resolution, continuous monitoring to evaluate potato varieties under rapidly changing climatic conditions. However, existing Exploratory Data Analysis (EDA) tools rarely integrate heterogeneous data streams – UAV imagery, IoT sensing, and field observations – into a unified, automated, and breeder-friendly workflow. This fragmentation limits breeders’ ability to detect stress, compare genotypes, and make timely, evidence-based decisions. We present a unified EDA system that consolidates multispectral UAV data, in-field IoT temperature measurements, historical agronomic records, and GIS-based spatial analyses into a single end-to-end architecture. The platform automates data acquisition, ETL processes, quality control, and relational storage while offering an interactive interface for flexible visualization and multi-scale analysis. The system was deployed over two growing seasons in Latvia using two potato varieties (Brasla and Imanta), 10 UAV missions per year, 4,800 multispectral images per season, and continuous five-minute sensor data. Field evaluations with professional breeders show that the system enhances data exploration, improves usability, and reduces information loss compared with traditional manual workflows. A qualitative assessment of the interface indicated greater user confidence and clearer access to key indicators. To address the absence of established benchmarks, we provide quantitative metrics demonstrating data quality, precision, automation performance, and integration reliability. Comparisons of our NDVI outputs with values reported in the literature further validate the robustness of the multispectral processing pipeline. Overall, this work introduces the first unified, breeder-oriented EDA framework that combines high-precision monitoring, automated multi-source integration, and interactive analytics. The results demonstrate how accessible digital tools can accelerate varietal selection and support more resilient and sustainable potato breeding under increasing environmental variability.
SN - 2376-5992
SN - 2376-5992
ER -