Comparison of Support Vector Machine and Neural Network Algorithm in Drone Detection System

With the increase in the number of drones, it is possible to have the danger of using drones illegally. It is crucial to detect adverse events or conditions so that security operators can obtain that information and situational identification of drones. This paper proposes two methods of classifying acoustic sensor data in a UAV detection system, using Support Vector Machine and Neural Network, that will be compared. This research shows that the accuracy achieved in predicting acoustic sensor data is 82.27% in the SVM method. The accuracy achieved is 90.58% for the NN method under the same input conditions and amount of training data. This comparation needs to do to choose the best accuracy in a public safety environment.