Published Paper Details:

Soil fertility is an important factor in assessing the quality of the soil since it shows how well the soil can support plant growth in agriculture.The quantity of nutrients in the soil can be quickly determined by using an Arduino board equipped with a soil sensor. The nutrient content of the soil can be quickly determined with an ESP32 equipped with a soil sensor. It is widely accepted that nitrogen, phosphorous, and potassium are essential constituents of nutrient sources.These elements should be measured to determine how much more nutrient content needs to be added to the soil to increase crop fertility.Soil fertility can be established through NPK sensors. By examining data on soil nutrient concentrations, we may determine whether the soil used to support plant production is rich in nutrients or lacking in them. There are several approaches to use a mass spectrogram or sensing element to find out how many nutrients are in the soil samples.However, with data only 60-70% genuine, the spectrum analysis approach is laborious.Due to a scarcity of data, a detailed comparison between the spectrum analysis approach and conventional wet chemistry procedures is required to ascertain the accuracy of the products. To detect soil nitrogen, phosphorus, and potassium, a soil NPK sensor should be used. Along with the application of a cheap, quick, straightforward, elevated, and portable soil NPK sensor. The plus side over a traditional detection method is the extremely short observation times required to gather exact data.This study compares and utilises different soil nutrient levels using machine learning and the kernel density estimation algorithm. One advising technological advancement that provides rational and sensible answers to the transformation taking place in many industries is the Internet of Things (IOT).Numerous research surveys and evaluations have been conducted, and a variety of methods have been used to apply IOT technology in agronomical fields.When it comes to quickly detecting plant health issues so that appropriate action may be taken, IOT can be very useful. It's a significant development in smart agriculture.Within this manuscript, we introduce a methodology for creating an automated framework capable of detecting crop damage in its early stages, which remain undetectable to the unaided eye. This method helps to prevent large losses in addition to saving a tonne of time and labour.The proposed method builds a recognition framework using soil-based detection devices, such as nitrogen, potassium, and phosphorus derived from plants. Sensor data is sent to the Arduino Cloud, which evaluates it and helps identify plant damage. In the years to come, the smart farming system will rely heavily on the internet of things.