SVM based target classification using RCS feature vectors

Bufler, Travis Dale; Narayanan, Ram M.; Dogaru, Traian

doi:10.1117/12.2176759

Cited by 4 publications

(5 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To get less complex systems, target classification from raw data offers a number of advantages. In this category, we find radar cross section (RCS) responses or micro-Doppler measurements that can be directly used to classify different objects, moving targets or human activities [10][11][12]. Recent works on object classification based on raw SAR measurements have even shown results that are only slightly inferior to pre-processed data, but at much lower computational costs [13,14].…”

Section: Introductionmentioning

confidence: 99%

Shape and orientation classification of objects based on their electromagnetic signatures using convolutional neural networks

Zaky,

Fortino,

Miramond

et al. 2024

Inverse Problems

View full text Add to dashboard Cite

This study addresses the classification of objects using their electromagnetic signatures with Convolutional Neural Networks (CNNs) trained on noiseless data. The singularity expansion method (SEM) was applied to establish a compact model that accurately represents the ultra-wideband scattered field of an object, independently of its orientation and observation angle. To perform the classification, we used a CNN associated with a noise-robust SEM technique to classify different objects based on their characteristic parameters. To validate this approach, we compared the performance of the classifier with and without SEM pre-processing of the scattered field for different noise levels and for object sizes not present in the training set. Moreover, we propose a procedure that determines the direction of the receiving antenna and orientation of an object based on the residues associated with each complex natural resonance. This classification procedure using pre-processed SEM data is promising, especially when generalizing to object sizes not included in the training set.

show abstract

Section: Introductionmentioning

confidence: 99%

Shape and orientation classification of objects based on their electromagnetic signatures using convolutional neural networks

Zaky,

Fortino,

Miramond

et al. 2024

Inverse Problems

View full text Add to dashboard Cite

show abstract

“…In fact, multiple data types are used for classification. Radar Cross Section (RCS) responses can be directly used as raw data, as in [5], [6], where authors used them to classify objects using Convolutional Neural Networks (CNNs) and SVM respectively. Moreover, Micro-Doppler measurements are used for the detection of humans and the classification of moving targets or human activities [7], [8].…”

Section: Introductionmentioning

confidence: 99%

“…In fact, multiple data types can be used for classification. Radar Cross Section responses can be directly used as raw data to classify objects (Bufler et al., 2015). Moreover, Micro‐Doppler measurements are used for the detection of humans and the classification of moving targets or human activities (Hadhrami et al., 2018; Yang et al., 2006).…”

Section: Introductionmentioning

confidence: 99%

“…Several algorithms are found in literature, that vary from simple models, like support vector machines (SVMs), decision trees (DTs), K-Nearest Neighbor or Naive Bayes, to more complex models like Artificial Neural Networks (ANNs) (Narayanan et al, 2017).In fact, multiple data types can be used for classification. Radar Cross Section responses can be directly used as raw data to classify objects (Bufler et al, 2015). Moreover, Micro-Doppler measurements are used for the detection of humans and the classification of moving targets or human activities (Hadhrami et al, 2018;Yang et al, 2006).…”

mentioning

confidence: 99%

See 1 more Smart Citation

Generalization Ability of Deep Learning Algorithms Trained Using SEM Data for Objects Classification

et al. 2022

View full text Add to dashboard Cite

This paper proposes a workflow to efficiently determine the material of spherical objects and the location of the receiving antenna relative to their position in bi-static measurements using supervised learning techniques. From a single observation, we compare classification performances resulting from the application of several classifiers on different data types: the Ultra-Wide Band scattered field in time and frequency domains and pre-processed data from the Singularity Expansion Method (SEM). Indeed, the resonances extracted using the SEM are aspect independent and therefore, are used to discriminate the objects. As for the residues, they depend upon the aspect angle and can hence be exploited to determine the observation angle. We construct 3 datasets to assess which one yields the highest accuracy while using the simplest and fastest classifiers. Hence, 80% of each dataset is used for training and the remaining 20% are used for testing. In a further step, we test with sphere sizes and data with several noisy levels that were not in the training datasets. Although SEM is noise sensitive, associating a robust feature extraction technique (Vector Fitting) with a simple but reliable classifier is promising, particularly when generalizing to data not included in the training set.

show abstract

“…Our previous work had explored how the spectral characteristics of indoor clutter compare to human beings by changing variables such as frequency, polarisation, and aspect angle. Initial work [7] in separating humans from indoor clutter has shown promising results, based on which a more thorough investigation is undertaken in this paper. Prior work has focused on the effects of furniture scattering on radar detection of humans [8–13] as well as wireless propagation within buildings [14, 15], and limited work on characterising the radar cross‐sections (RCSs) of furniture [16].…”

Section: Introductionmentioning

confidence: 99%