Uso de sensores remotos en la agricultura: aplicaciones en el cultivo del banano

Abstract

Introduction. Remote sensors offer the ability to observe an object without being in contact with it. They are widely used in agricultural applications and have large development potential in banana (Musa AAA) plantations. During the past decades, the research in remote sensing and agriculture has increased through the availability of high-resolution satellite images (spatial, spectral, and temporal) and the use of remotely piloted vehicles that generate base information for research. Objective. To carry out a general review on the applications of the use of remote sensors for banana plantations in three specific aspects: determination of the cultivation area, productivity estimation, and disease diagnosis. Development. The extension of land covered by commercial banana plantations can be detected visually or easily by means of remote image classifications, such as the Synthetic Aperture Radar (SAR) sensor, which hve resulted in classification accuracies of around 95%. This is due to the high backscattering of the large leaves of the plant. However, the studies on productivity are scarce for banana cultivation and have been limited to the use of vegetation index, showing poor results in their correlations. As for the identification of diseases, work has been done on the main diseases affecting production with correlation levels above 90 % for some diseases. Conclusion. This review shows that banana plantations can be detected through the use of remote sensors and, likewise, these allow the identification of the main diseases in the crop. However, the results obtained to determine productivity are scarce and with little precision.

Plant phenotyping relevance

バナナ作物の生産性推定と病害診断におけるリモートセンシング手法をレビューしており、植物状態の取得・推定方法が中心である。

Code and data availability

This is a bibliographic review of remote sensing applications in banana cultivation. The supplied blocks contain only the article text and its reference list; all URLs correspond to cited prior works, author ORCID profiles, or institutional statistics pages. No public phenotype/trait dataset, plant/sensor imagery, or作者

No evidence-backed public reproduction asset is currently recorded.