IoT Driven AI-Based Agriculture Recommendation Model for Fertilizer

Fecha de publicación: 18/02/2022
Fuente: WIPO Agriculture Portada
According to a survey conducted by the Food and Agriculture Organization, the global population is predicted to rise by another two billion people by 2050, but farmland is only expected to grow by 5%. As a result, to increase agriculture productivity, smart and efficient farming techniques are required. One of the most important tools for agriculture improvement is fertiliser assessment. As an alternative to collecting and processing farm data, several new technologies and innovations are being applied in agriculture. With the Internet of Things (IoT) established as a viable instrument for automating and decision-making in the domain of agriculture, the rapid growth of wireless sensor networks has prompted the design of low-cost and tiny sensor devices. For the assessment of agricultural fertiliser, this study presents an expert system that integrates sensor networks with Artificial Intelligence systems such as Neural Networks and Multi-Layer Perceptron (MLP). This proposed system will assist farmers in evaluating agriculture fertiliser for growing by categorising it into four decision classes: more suitable, suitable, moderately suitable, and unsuitable. This evaluation is based on the data gathered from the numerous sensor devices that are utilised to train the system. When compared to other current models, the results produced using MLP with four hidden layers are determined to be effective for the multiclass classification system. After each cultivation, this trained model will be used to evaluate future assessments and classify the land-based fertilizer. Fertilizer recommendations can be made based on fertilizer, crop, and geographical information. This section recommends appropriate crops and fertilizer for each crop.