Abstract
AI has recently become a powerful tool for analyzing data where the informational value is not known, helping to identify which parts of the measured data contain relevant information and which do not. For example, AI was used to discover 300 new ancient inscriptions on the Nazca Plateau in Peru. This was achieved through the analysis of multispectral aerial imagery, from which meaningful patterns were previously undetectable by humans until AI was applied, leading to this groundbreaking discovery. We have used AI to analyze hyperspectral and multispectral data, and here we present how these new approaches are aiding in the detection of both biotic and abiotic stress. However, traditional analytical methods, which typically rely on vegetation index calculations, only use a small portion of the available spectral data and often cannot be tailored to the specific needs of each unique study. By leveraging machine learning, we can efficiently utilize all of this information to flexibly assess plant health, metabolite content, fruit quality and ripening, and even contribute to the development of future medical solutions.
