A review of Machine Learning approach for ground penetrating radar applications

Machine learning (ML) is a branch of artificial intelligent in which algorithms learn relationships in data. ML can be applied in predictive sense or to investigate internal relationships of dataset. The ability to give promising results bring ML been applied in various application such as imaging, signals processing, data mining, and many more. In this paper, the ML approach for Ground Penetrating Radar application is reviewed. Nowadays, Ground Penetrating Radar have some issues of accuracy of localization and the image processing due to the noise and unwanted signal from the underground. Therefore, some of the smart learning technique is proposed especially to remove the clutter signals. A comparison of ML technique such as linear regression, logistic regression, KNN, support vector machine and etc for clutter issues is presented in this paper. The most suitable technique for in GPR applications in order to solve the clutter issues is proposed