Object Identification Based on the Perturbation Analysis of the Sound Field in The Room Environment

Object identification in the room environment is a key technique in many advanced engineering applications such as the unidentified object recognition in security surveillance, human identification and barrier recognition for AI robots. In this paper, a novel object identification technique based on the room acoustics knowledge is presented. This technique is proposed based on the fact that each object has specific influence pattern to the indoor sound fields, regarding their geometry and acoustic characteristics. The influence is expressed by the variations of the room impulse response (RIR) which represents the wave propagating path from the source to the receiver. The technique is implemented under the classical framework of pattern recognition and its effectiveness is explored through two experiments. Experimental results indicate that the identification accuracy is strong related to the type of the data features and the locations of the objects in the step of data collection. The accuracy can reach up to 92.3% under a proper condition, which shows that the technique has potential to be an alternative method in the field of object identification.