No abstract

As a non-invasive and remote sensor, the Laser Doppler Vibrometer (LDV) has found a broad spectrum of applications in various areas such as civil engineering, biomedical engineering, and even security and restoration within art museums. LDV is an ideal sensor to detect threats earlier and provide better protection to society, which is of utmost importance to military and law enforcement institutions. However, the use of LDV in situational surveillance, in particular vehicle classification, is still in its infancy due to the lack of systematic investigations on its behavioral properties. In this work, as a result of the pilot project initiated by Air Force Research Laboratory, the innate features of LDV data from many vehicles are examined, beginning with an investigation of feature differences compared to human speech signals. A spectral tone-pitch vibration indexing scheme is developed to capture the engine’s periodic vibrations and the associated fundamental frequencies over the vehicles’ surface. A two-layer feed-forward neural network with 20 intermediate neurons is employed to classify vehicles’ engines based on their spectral tone-pitch indices. The classification results using the proposed approach over the complete LDV dataset collected by the project are exceedingly encouraging; consistently higher than 96% accuracies are attained for all four types of engines collected from this project.

In our previous studies, vehicle surfaces' vibrations caused by operating engines measured by Laser Doppler Vibrometer (LDV) have been effectively exploited in order to classify vehicles of different types, e.g., vans, 2-door sedans, 4-door sedans, trucks, and buses, as well as different types of engines, such as Inline-four engines, V-6 engines, 1-axle diesel engines, and 2-axle diesel engines. The results are achieved by employing methods based on an array of machine learning classifiers such as AdaBoost, random forests, neural network, and support vector machines. To achieve effective classification performance, we seek to find a more reliable approach to pick authentic vibrations of vehicle engines from a trustworthy surface. Compared with vibrations directly taken from the uncooperative vehicle surfaces that are rigidly connected to the engines, these vibrations are much weaker in magnitudes. In this work we conducted a systematic study on different types of objects. We tested different types of engines ranging from electric shavers, electric fans, and coffee machines among different surfaces such as a white board, cement wall, and steel case to investigate the characteristics of the LDV signals of these surfaces, in both the time and spectral domains. Preliminary results in engine classification using several machine learning algorithms point to the right direction on the choice of type of object surfaces to be planted for LDV measurements.

No abstract

No abstract