Plant nutritional deficiency detection: a survey of predictive analytics approaches

Detecting plant nutritional deficiencies is crucial in agriculture, as these deficiencies directly impact productivity, food security, and the environment. Conventional methods for assessing plant nutritional content, such as soil testing and leaf tissue analysis, are time-consuming and expensive. Over the past decade, researchers have been working on automating this process using predictive analytics. This study discusses the importance of essential nutrients in plants, their visual symptoms, and various methods for assessing nutritional deficiencies. This study categorizes current research into two categories based on the type of data used for prediction: visible spectral images and multispectral/hyperspectral images. This classification offers valuable insight into the strengths and limitations of each, thereby shedding light on their potential applications. This study examined the challenges and possible solutions in automating the detection of nutritional deficiencies. It highlights the need for scalable and accessible solutions and emphasizes human–machine collaboration for precision and interpretability. By addressing these difficulties and leveraging predictive analytics capabilities, substantial progress can be made in detecting nutritional deficiencies, contributing to improving agricultural practices.