Abstract:An image classification technique, which has recently been introduced for visual pattern recognition, is successfully applied for human gait classification based on radar Doppler signatures depicted in the time-frequency domain. The proposed method has three processing stages. The first two stages are designed to extract Doppler features that can effectively characterize human motion based on the nature of arm swings, and the third stage performs classification. Three types of arm motion are considered: free-a… Show more
“…The differences between the armed and unarmed case is easily noticeable by eye for both monostatic and bistatic data, in particular the fact that in the armed case the micro-Doppler signature is more compressed around the main torso contribution at around 20 Hz, without the peaks associated to the free swinging movements of the limbs. This is valuable information, as it has been shown that limited and confined limbs movement may be linked to people carrying potentially hostile objects, or to the presence of injured people or hostages [12,13]. These differences between spectrograms in the armed and unarmed cases will be numerically quantified and converted into features to use as input to a classifier.…”
“…The differences between the armed and unarmed case is easily noticeable by eye for both monostatic and bistatic data, in particular the fact that in the armed case the micro-Doppler signature is more compressed around the main torso contribution at around 20 Hz, without the peaks associated to the free swinging movements of the limbs. This is valuable information, as it has been shown that limited and confined limbs movement may be linked to people carrying potentially hostile objects, or to the presence of injured people or hostages [12,13]. These differences between spectrograms in the armed and unarmed cases will be numerically quantified and converted into features to use as input to a classifier.…”
“…In the latter case the person had both arms free to swing as in natural walking. The use of micro-Doppler to distinguish between free and confined arms swinging may be of interest, as confined arms and reduced limbs movement could be related to people carrying potentially hostile objects, or to the presence of hostages or injured people [25][26][27]. Walking on the spot removes the main Doppler shift contribution from the micro-Doppler signatures, and ensures that targets remain within the main beam of the transmitting and receiving antennas during the recording.…”
Section: Data Collection and Experimental Setupmentioning
This paper discusses the analysis of multistatic micro-Doppler signatures and related features to distinguish and classify unarmed and potentially armed personnel. The application of radar systems to distinguish different motion types has been previously proposed and this work aims to further investigate the applicability of this in more scenarios. Real data have been collected using a multistatic radar system in a series of experiments involving several individuals performing different movements. Changes in classification accuracy as a function of different aspect angle between the direction in which the target faces and the line-of-sight of the radar nodes are analysed. Multiple data fusion methodologies are proposed, showing that significant improvement of the classification accuracy can be achieved when using separate classification at each node followed by a voting procedure to reach the final decision. This is beneficial especially at those aspect angles for which micro-Doppler detection is less favourable.
“…The use of micro-Doppler to distinguish between free and confined arms swinging may be of interest, as confined arms and reduced limbs movement could be related to people carrying potentially hostile objects or to the presence of hostages or injured people. This type of analysis can exploit positive and negative micro-Doppler caused by the arms and its periodicity, and was the core contribution of these works [11][12][13] and investigated in our previous paper [14]. These observations seem to disagree with those proposed in [15], where it is claimed that carrying objects with one or both hands does not change the micro-Doppler signature for a walking person.…”
Abstract-Human micro-Doppler radar signatures have been investigated to classify different types of activities and to identify potential armed personnel in the context of security and surveillance applications. In this paper the use of multistatic micro-Doppler signatures to distinguish between unarmed and armed personnel moving is described. The effect of polarimetry on the classification accuracy is evaluated. Real radar data from a multistatic radar (NetRAD) has been analyzed as part of this work. Suitable features are extracted from the spectrograms generated from the data and then used as input to a classifier. The impact of polarization diversity on the classification performance is investigated, in particular the use of co-polarized or cross-polarized data or their multistatic combination.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.