Urban Objects Classification With an Experimental Acoustic Sensor Network

Abstract

This paper proposes feature extraction methods for object classification with passive acoustic sensor networks deployed in (sub-)urban environments. We analyzed the emitted acoustic signals of three object classes: guns (muzzle blast), vehicles (running piston engine) and pedestrians (several footsteps). Based on the conducted analysis, methods are developed to extract features that are related to the physical nature of the objects. In addition, localization methods are developed (e.g. pseudo-matched-filter), because the object location is required for one of the feature extraction methods. As a result, we developed a proof-of-concept system to record and extract discriminative acoustic features. The performance of the features and the final classification are assessed with real measured data of the three object classes within sub-urban environment.

A comprehensive review of fruit and vegetable classification techniques

“…Sensors ranging from black and white (B/W) cameras to non-visual sensors such as acoustic and tactile sensors have been used for classification of fruit and vegetable, but not all sensors are equally suitable for all applications. As evident from [9,10,11,17,18,19,82] both acoustic and tactile sensors are less suitable for nondestructive classification and recognition. These sensors either need physical contact or excitation of the fruit or vegetable for data acquisition.…”

“…The basic properties of absorption, reflection and refraction of acoustic signals have been used for classification of fruit and vegetable. Acoustic signals have been used for quality assessment of fruit and vegetable by measuring their elasticity as a function of hydration content in their tissues [19,26,27,82]. In the acoustic analysis a fruit is excited by a physical impact to produce an acoustic wave used to measure the elastic modulus to confirm the firmness and hence freshness.…”