Distinguished Study of Various Materials Like Metal Mine/ Rock from Underground Data of Unmanned Vehicle Using Machine Learning Techniques

Abstract

The detection of minerals (mine) or rock would have been very complicated through the enlargement of Sound Navigation Ranging (SONAR) technique that passes on particular parameters to be proficient to distinguish the obstruction or the surface is a mine or rock. Machine learning techniques pay greater understanding of classification and regression issues correlated with many domain technologies and industries through statistical analysis. In this novel study, machine learning algorithms such as Gaussian Naïve Bayes and Decision tree classifier which identifies the classification that may either rock or mine or some other creature or other kind of body. Moreover, the experiment is a straightforward comes up with a machine learning training to rate the underwater as mine or rock accomplished on a large, extremely spatial, and difficult SONAR dataset. The accuracy for both training and testing samples can be evaluated using classifiers for separating rock or mine in under water acoustics through SONAR. The decision tree classifiers outperform the results by achieving high accuracy with less prediction time which enhanced the performance of the model. Machine learning act as a huge role function to improve the efficiency of natural resource detection underwater, and will appear to be perfect for the near future.