Smart-farming:IoT Based Crop Prediction System using Machine Learning Techniques
Abstract
The growth of the human population over the decades has led to a constant rise in demand for agricultural products. The smart farming technique maximizes the efficiency of crop production. Smart techniques enhance the agricultural sector by providing advanced technology that incorporates internet of things (IoT), big data, and cloud. Crop selection plays a major role to increase the yield and provide farmers with good profits. The crop growth is based on various factors that include temperature, soil moisture, air humidity, rainfall, light, and pH value of the soil. Machine learning techniques are applied to the environment conditions and prediction of crop is done on the basis of the obtained result. The proposed system uses sensor technology to monitor the farm and predict the crop using the Support Vector Machine(SVM) algorithm. The system contains a pH sensor, a soil moisture sensor, and a DHT11 sensor which reads the data of the field. The collected data from the IoT equipment is sent to the ThingSpeak server. On the collected data we apply the SVM classification technique to predict the crop suitable for the farm. SVM classifier provides good accuracy and faster prediction of the classes. This system would help farmers to improve productivity and reduce wastage by enabling better management of the yield through smart farming..