Panama disease in banana plantations

Multispectral analysis for the detection of panama disease in banana plantations

In this study, a multispectral analysis was conducted in a banana plantation to identify areas affected by panama disease (Fusarium oxysporum f. sp. cubense), one of the most destructive diseases in banana crops. Using drones equipped with multispectral cameras, strategic flights were carried out to capture images in different spectral bands, enabling the generation of vegetation indices such as NDVI and NDRE, which are essential for assessing the physiological condition of the plants.

One of the main challenges during the analysis was the structural complexity of the banana plantation, characterized by the presence of mother plants, daughters, and grandchildren. This configuration generates variability in reflectance due to leaf overlap and projected shadows, which can distort the interpretation of spectral indices. To mitigate these difficulties, data from the Digital Surface Model (DSM) and the Digital Terrain Model (DTM) were integrated, allowing differentiation of plant height and filtering of shadow effects and foliage density.

Thanks to this combined methodology, it was possible to accurately map the areas affected by Panama Disease, identifying zones with early vegetative stress and facilitating decision-making for the phytosanitary management of the crop. This approach highlighted the importance of using topographic models in conjunction with multispectral analysis to achieve more precise results in structurally complex crops such as banana plantations.