Smart Pesticide Detection System Using IoT and ML

Initially, we insert any fruit in front of the color sensor for scanning purposes. To measure pH, we use a pH sensor, and to detect if the fruit generates any gas, we use an MQ135 sensor. In the machine learning process, a dataset is created using the Random Forest algorithm. This dataset includes various properties and measurements such as color (from the color sensor), temperature, humidity, and gas emission values. These features are used to determine whether a fruit is of good quality or not. By comparing the current fruit’s values with the pre-defined dataset, the system evaluates its quality. If the fruit meets the best or pre-defined criteria/ranges, the ML model detects it as good; otherwise, a buzzer activates to indicate that the fruit is not good.