摘要

An approach to recognizing two objects using CTFM ultrasonics is discussed in this paper. Extracting features from each object echoes power spectral density is detailed, and a method of object recognition using feature vectors and the Mahalanobis distance is presented. The results of this research indicate that it is possible to recognize simple geometry objects like a table and chair, which produce little information in their echoes. It is also possible to differentiate between varying perspectives of the same object.

  • 出版日期2012-6