- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
A New Algorithm for Blood Flow Measurement Based on The Doppler Flow Spectrogram
摘要： The blood ?ow is traditionally obtained by multiplying the cross sectional area of the blood vessel and the average blood speed in the cross section, or is given by the integral of the product of the cross section and blood velocity of each element. However, both methods are affected greatly by the measurement precision of the area and velocity. A new algorithm, which is based on the Doppler blood ?ow spectrogram, is proposed to measure the blood ?ow in this paper. In the algorithm, the blood ?ow is calculated according to the double integral of a Doppler blood ?ow spectrogram. To verify the feasibility of the proposed algorithm, experiments have been performed on the Doppler blood-mimicking system KS205D?1 using the SonixTouch ultrasonic system. In addition, linear-regression analysis is carried out to observe the correlation factors between the experimental values and real values of different ?ow rates. Experimental results show that the calculated values and real values correlate signi?cantly (r > 0.969, P < 0.0000001). Experimental results both on males and females also veri?ed the proposed algorithm (r > 0.915, P < 0.00053). Hence the proposed algorithm is proven effective for relative mean blood ?ow measurement. Due to the special structure of the human brain, it is dif?cult to measure the cross sectional area of blood vessel with ultrasound imaging. In this algorithm, there is no need to measure the cross sectional area of the blood vessel. Therefore, the proposed algorithm has the potential to be a new method for clinical ultrasonic blood ?ow measurement, especially cerebral blood ?ow measurement.
关键词： Blood ?ow measurement,Doppler blood-mimicking system,Doppler spectrogram,linear-regression analysis
更新于2025-09-23 15:22:29
-
[IEEE 2018 15th European Radar Conference (EuRAD) - Madrid, Spain (2018.9.26-2018.9.28)] 2018 15th European Radar Conference (EuRAD) - Deep Learning based Human Activity Classification in Radar Micro-Doppler Image
摘要： A convolutional neural network (CNN) based deep learning (DL) approach to classify human activities in micro-Doppler spectrogram of radar is investigated. MOCAP dataset, from Carnegie Mellon University, is used for spectrogram simulation. Seven activities are classified with the proposed CNN network. Our network outperforms several previously published DL-based approaches. To understand the network’s impact on classification performance, we investigate some key parameters of the proposed network. Experiment result demonstrates that a deeper network does not necessarily result in a higher accuracy. We also examine the network size and the number of output feature maps to find out their impact on the result.
关键词： Deep Learning,Convolutional Neural Network,Human Activity Classification,Micro-Doppler Spectrogram,Radar image
更新于2025-09-23 15:21:21
-
[IEEE 2019 IEEE 46th Photovoltaic Specialists Conference (PVSC) - Chicago, IL, USA (2019.6.16-2019.6.21)] 2019 IEEE 46th Photovoltaic Specialists Conference (PVSC) - Stability of Tin-Lead Halide Perovskite Solar Cells
摘要： Envelope representations such as the auditory or traditional spectrogram can be defined by the set of envelopes from the outputs of a filterbank. Common envelope extraction methods discard information regarding the fast fluctuations, or phase, of the signal. Thus, it is difficult to invert, or reconstruct a time-domain signal from, an arbitrary envelope representation. To address this problem, a general optimization approach in the time domain is proposed here, which iteratively minimizes the distance between a target envelope representation and that of a reconstructed time-domain signal. Two implementations of this framework are presented for auditory spectrograms, where the filterbank is based on the behavior of the basilar membrane and envelope extraction is modeled on the response of inner hair cells. One implementation is direct while the other is a two-stage approach that is computationally simpler. While both can accurately invert an auditory spectrogram, the two-stage approach performs better on time-domain metrics. The same framework is applied to traditional spectrograms based on the magnitude of the short-time Fourier transform. Inspired by human perception of loudness, a modification to the framework is proposed, which leads to a more accurate inversion of traditional spectrograms.
关键词： short-time Fourier transform,auditory spectrogram,Spectrogram inversion,gradient methods
更新于2025-09-23 15:19:57
-
[IEEE 2019 6th International Conference on Systems and Informatics (ICSAI) - Shanghai, China (2019.11.2-2019.11.4)] 2019 6th International Conference on Systems and Informatics (ICSAI) - Laser Strip Center Extraction Methodology for the Detection of Weld Seam
摘要： Envelope representations such as the auditory or traditional spectrogram can be defined by the set of envelopes from the outputs of a filterbank. Common envelope extraction methods discard information regarding the fast fluctuations, or phase, of the signal. Thus, it is difficult to invert, or reconstruct a time-domain signal from, an arbitrary envelope representation. To address this problem, a general optimization approach in the time domain is proposed here, which iteratively minimizes the distance between a target envelope representation and that of a reconstructed time-domain signal. Two implementations of this framework are presented for auditory spectrograms, where the filterbank is based on the behavior of the basilar membrane and envelope extraction is modeled on the response of inner hair cells. One implementation is direct while the other is a two-stage approach that is computationally simpler. While both can accurately invert an auditory spectrogram, the two-stage approach performs better on time-domain metrics. The same framework is applied to traditional spectrograms based on the magnitude of the short-time Fourier transform. Inspired by human perception of loudness, a modification to the framework is proposed, which leads to a more accurate inversion of traditional spectrograms.
关键词： Spectrogram inversion,gradient methods,short-time Fourier transform,auditory spectrogram
更新于2025-09-23 15:19:57
-
Griffin-Lim Like Phase Recovery via Alternating Direction Method of Multipliers
摘要： Recovering a signal from its amplitude spectrogram, or phase recovery, exhibits many applications in acoustic signal processing. When only an amplitude spectrogram is available and no explicit information is given for the phases, the Grif?n–Lim algorithm (GLA) is one of the most utilized methods for phase recovery. However, GLA often requires many iterations and results in low perceptual quality in some cases. In this paper, we propose two novel algorithms based on GLA and the alternating direction method of multipliers (ADMM) for better recovery with fewer iteration. Some interpretation of the existing methods and their relation to the proposed method are also provided. Evaluations are performed with both objective measure and subjective test.
关键词： STFT-based speech synthesis,spectrogram consistency,short-time Fourier transform (STFT),Non-convex optimization,phaseless spectrogram inversion
更新于2025-09-09 09:28:46
-
[ASME ASME 2018 International Technical Conference and Exhibition on Packaging and Integration of Electronic and Photonic Microsystems - San Francisco, California, USA (Monday 27 August 2018)] ASME 2018 International Technical Conference and Exhibition on Packaging and Integration of Electronic and Photonic Microsystems - Assessment of Damage Progression in Automotive Electronics Assemblies Subjected to Temperature and Vibration
摘要： Electronics in automotive underhood environments is used for a number of safety critical functions. Reliable continued operation of electronic safety systems without catastrophic failure is important for safe operation of the vehicle. There is need for prognostication methods, which can be integrated, with on-board sensors for assessment of accrued damage and impending failure. In this paper, leadfree electronic assemblies consisting of daisy-chained parts have been subjected to high temperature vibration at 5g and 155°C. Spectrogram has been used to identify the emergence of new low frequency components with damage progression in electronic assemblies. Principal component analysis has been used to reduce the dimensionality of large data-sets and identify patterns without the loss of features that signify damage progression and impending failure. Variance of the principal components of the instantaneous frequency has been shown to exhibit an initial damage progression, increasing trend during the attaining a maximum value and decreasing prior to failure. The unique behavior of the instantaneous frequency over the period of vibration can be used as a health-monitoring feature for identifying the impending failures in automotive electronics. Further, damage progression has been studied using Empirical Mode Decomposition (EMD) technique in order to decompose the signals into Independent Mode Functions (IMF). The IMF’s were investigated based on their kurtosis values and a reconstructed strain signal was formulated with all IMF’s greater than a kurtosis value of three. PCA analysis on the reconstructed strain signal gave better patterns that can be used for prognostication of the life of the components.
关键词： high temperature vibration,prognostication,Empirical Mode Decomposition,spectrogram,kurtosis,automotive electronics,principal component analysis
更新于2025-09-09 09:28:46