An efficient and robust multi-frame image super-resolution reconstruction using orthogonal Fourier-Mellin moments

Singh, Chandan; Aggarwal, Ashutosh

doi:10.1016/j.displa.2017.06.002

Cited by 6 publications

(3 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To improve the detection performance, a time-domain homogenization fusion algorithm is proposed, i.e. by averaging the multi-frame sequences, so that the target energy is greatly enhanced while the noise energy is a little increased due to the high correlation of the target signal between successive frames, which effectively improves the signal-to-noise ratio [19].…”

Section: Time-domain Homogenization Fusion Algorithmmentioning

confidence: 99%

Patient behavior detection based on time-frequency fusion of FMCW radar

Xu¹,

Zhang²,

Chen³

et al. 2022

Meas. Sci. Technol.

View full text Add to dashboard Cite

This paper proposes a novel time-frequency feature fusion method to recognise patients’ behaviours based on the Frequency Modulated Continuous Wave (FMCW) radar system, which can locate patients as well as recognise their current actions and thus is expected to solve the shortage of medical staff caused by the novel coronavirus pneumonia(COVID-19).To recognise the patient’s behaviour, the FMCW radar is utilised to acquire point clouds reflected by the human body, and the micro-Doppler spectrogram is generated by human motion. Then features are extracted and fused from the time-domain information of point clouds and the frequency-domain information of the micro-Doppler spectrogram respectively. According to the fused features, the patient’s behaviour is recognised by a Bayesian optimisation algorithm which selects optimal hyper-parameters, i.e., the number of random forest decision trees, the depth of leaves, and the number of features. The experimental results show that an average accuracy of 99.3% can be achieved by using the time-frequency fusion with the Bayesian optimisation random forest model to recognise six actions.

show abstract

Section: Time-domain Homogenization Fusion Algorithmmentioning

confidence: 99%

Patient behavior detection based on time-frequency fusion of FMCW radar

Xu¹,

Zhang²,

Chen³

et al. 2022

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…In this Letter, 12 different moment methods, namely Legendre moments (LM) [1], Chebyshev moments (CM) of first (CM#1) and second kind (CM#2) [2], Gegenbauer moments (GM) [3], Jacobi moments (JM) [4], Krawchouck moments (KM) [5], Zernike moments (ZM) [6], pseudo ZM (PZM) [7], Fourier–Merlin moments (FMM) [8], Chebyshev–Fourier moments (CFM) [9], radial harmonic Fourier moments (RHFM) [10] and radial CM (RCM) [6] are used in the moving stationary target acquisition and recognition (MSTAR) images [11] for feature extraction. Although the number of features extracted can be arranged by the order and repetition parameters of the moments [6], based on the information content, there is a limit to extract independent and distinctly informative features.…”

Section: Introductionmentioning

confidence: 99%

“…Owing to highresolution characteristics, different methods with dimensionality reduction capabilities are adopted for feature extraction in the literature. Moments are introduced in the literature as effective methods to extract features that are scale, translation and rotational invariant with minimum information redundancy [1][2][3][4][5][6][7][8][9][10].…”

mentioning

confidence: 99%

Entropy‐score‐based feature selection for moment‐based SAR image classification

2018

View full text Add to dashboard Cite

Entropy score is defined to be a metric indicating the class separation, which is then used in feature selection process to improve the momentbased classification performance of synthetic aperture radar (SAR) images. Feature extraction is performed over each SAR image by employing different moment methods to enrich the feature space before feature selection. Fusing all the features coming from different moment methods into a single vector is not feasible since the vector will have high dimensionality and embedded redundancies due to correlations among features. To reduce the dimensionality of feature space and increase the discrimination capability of the feature vector, a unique approach based on entropy score selecting top k methods for each feature coefficient is proposed. Experimental results verify the improvement of the accuracy on SAR image classification over the state-of-the-art.

show abstract