System for the Generation and Georeferenced Visualization of Agricultural Crop Datasets Using Imagro Hybrid Platform

Abstract

The agriculture sector in Ecuador faces significant challenges in its digital transformation process, primarily due to the lack of accessible tools for the capture and management of georeferenced crop data. In this context, this study proposes Imagro, a hybrid platform comprised of a mobile application and a web application, designed to capture, store, visualize, and classify crop images with integrated spatial metadata. This solution incorporates lightweight machine learning models, specifically MobileNet V1 with TensorFlow.js, which enables real-time automatic crop classification from web browsers, without the need for specialized infrastructure. The research employs a hybrid methodological approach that combines documentary techniques, such as systematic literature reviews and comparative analysis of tools. with experimental techniques including evolutionary prototyping, functional tests, and usability assessment. The inductive method was used to analyze user experience through direct observation and the application of the System Usability Scale (SUS) questionnaire, and analytical methods were used to study the system architecture, the performance of the model, and its main functionalities. The dataset used was built from open repositories and expert contributions, and the Imagro platform has been evaluated by users in agricultural and academic environments. The results show an SUS score of 78.2 and a 98% success rate in task execution, confirming the potential of Imagro as an accessible and functional digital tool to strengthen precision agriculture, boost applied research, and support agricultural education in rural contexts with limited connectivity.

Received: 10 August 2025 | Revised: 31 December 2025 | Accepted: 4 February 2026

Conflicts of Interest

The authors declare that they have no conflicts of interest to this work.

Data Availability Statement

The data that support the findings of this study are openly available in GitHub at https://github.com/Darling-P11/imagro_web.

Author Contribution Statement

Lucrecia Llerena: Methodology, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization. Nancy Rodríguez: Software, Supervision, Project administration, Validation, Investigation. Kevin Ponce: Methodology, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization, Software, Supervision, Project administration, Validation, Investigation.