- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
An image thresholding approach based on Gaussian mixture model
摘要： Image thresholding is an important technique for partitioning the image into foreground and background in image processing and analysis. It is difficult for traditional thresholding methods to get satisfactory performance on the noisy and uneven grayscale images. In this paper, we propose an image thresholding approach based on Gaussian mixture model (GMM) to solve this problem. GMM assumes that image is a mixture of two unknown parameters’ Gaussian distributions, which corresponds to foreground and background, respectively. Based on this assumption, we adopt expectation maximization algorithm with a simple initialization strategy to estimate the statistical parameters and utilize Bayesian criteria to generate the binary map. Furthermore, we calculate the posterior probabilities in consideration of neighborhood effect to achieve good performance on noisy and uneven grayscale images. Experimental results conducted on the synthetic and real images demonstrate the effectiveness of the proposed method.
关键词： Image thresholding,Gaussian mixture model,EM algorithm,Neighborhood information
更新于2025-09-23 15:23:52
-
[IEEE 2018 Signal Processing: Algorithms, Architectures, Arrangements, and Applications (SPA) - Poznan, Poland (2018.9.19-2018.9.21)] 2018 Signal Processing: Algorithms, Architectures, Arrangements, and Applications (SPA) - Hardware implementation of the Gaussian Mixture Model foreground object segmentation algorithm working with ultra-high resolution video stream in real-time
摘要： In this paper a hardware implementation of the Gaussian Mixture Model algorithm for background modelling and foreground object segmentation is presented. The proposed vision system is able to handle video stream with resolution up to 4K (3840x2160 pixels) and 60 frames per second. Moreover, the constraints caused by memory bandwidth limit are also discussed and a few different solutions to tackle this issue have been considered. The designed modules have been verified on the ZCU102 development board with Xilinx Zynq UltraScale+ MPSoC device. Additionally, the computing performance and power consumption have been estimated.
关键词： FPGA,4K video,background modelling,real-time processing,GPU,Gaussian Mixture Model,foreground object segmentation
更新于2025-09-23 15:23:52
-
Unsupervised Segmentation of Spectral Images with a Spatialized Gaussian Mixture Model and Model Selection
摘要： In this article, we describe a novel unsupervised spectral image segmentation algorithm. This algorithm extends the classical Gaussian Mixture Model-based unsupervised classification technique by incorporating a spatial flavor into the model: the spectra are modelized by a mixture of K classes, each with a Gaussian distribution, whose mixing proportions depend on the position. Using a piecewise constant structure for those mixing proportions, we are able to construct a penalized maximum likelihood procedure that estimates the optimal partition as well as all the other parameters, including the number of classes. We provide a theoretical guarantee for this estimation, even when the generating model is not within the tested set, and describe an efficient implementation. Finally, we conduct some numerical experiments of unsupervised segmentation from a real dataset.
关键词： Spectral images,Gaussian Mixture Model,Model selection,Spatial information,Unsupervised segmentation
更新于2025-09-23 15:22:29
-
An Efficient and Robust Iris Segmentation Algorithm Using Deep Learning
摘要： Iris segmentation is a critical step in the entire iris recognition procedure. Most of the state-of-the-art iris segmentation algorithms are based on edge information. However, a large number of noisy edge points detected by a normal edge-based detector in an image with specular reflection or other obstacles will mislead the pupillary boundary and limbus boundary localization. In this paper, we present a combination method of learning-based and edge-based algorithms for iris segmentation. A well-designed Faster R-CNN with only six layers is built to locate and classify the eye. With the bounding box found by Faster R-CNN, the pupillary region is located using a Gaussian mixture model. Then, the circular boundary of the pupillary region is fit according to five key boundary points. A boundary point selection algorithm is used to find the boundary points of the limbus, and the circular boundary of the limbus is constructed using these boundary points. Experimental results showed that the proposed iris segmentation method achieved 95.49% accuracy on the challenging CASIA-Iris-Thousand database.
关键词： Iris segmentation,Faster R-CNN,Gaussian mixture model,Boundary point selection,Deep learning
更新于2025-09-23 15:22:29
-
[IEEE 2019 IEEE 8th International Conference on Advanced Optoelectronics and Lasers (CAOL) - Sozopol, Bulgaria (2019.9.6-2019.9.8)] 2019 IEEE 8th International Conference on Advanced Optoelectronics and Lasers (CAOL) - The Fast Modification of Evolutionary Bioinspired Cat Swarm Optimization Method
摘要： In this paper, we propose a novel probabilistic method for the task of text-independent speaker identification (SI). In order to capture the dynamic information during SI, we design super-mel-frequency cepstral coefficients (MFCCs) features by cascading three neighboring MFCCs frames together. These super-MFCC vectors are utilized for probabilistic model training such that the speaker’s characteristics can be sufficiently captured. The probability density function (PDF) of the aforementioned super-MFCCs features is estimated by the recently proposed histogram transform (HT) method. To recede the commonly occurred discontinuity problem in multivariate histograms computing, more training data are generated by the HT method. Using these generated data, a smooth PDF of the super-MFCCs vectors is obtained. Compared with the typical PDF estimation methods, such as Gaussian mixture model, promising improvements have been obtained by employing the HT-based model in SI.
关键词： Speaker identification,Gaussian mixture model,mel-frequency cepstral coefficients,histogram transform model
更新于2025-09-23 15:21:01
-
Enhanced state estimation and bad data identification in active power distribution networks using photovoltaic power forecasting
摘要： In view of the problems of insu?cient real-time measurements in active distribution networks, a state estimation method for active distribution networks is proposed based on the forecasting of photovoltaic (PV) power generation. First, the extreme learning machine (ELM) enhanced by the genetic algorithm (GA) is used to forecast the PV power generation. Second, the Gaussian mixture model (GMM) is used to model the forecasting error. The weighted mean of the forecasting error is used to correct the forecasting value of the PV power generation, and the weighted variance of the forecasting error is used as the basis for setting the pseudo measurement weight. Finally, the real-time measurements collected by the supervisory control and data acquisition (SCADA) system, the forecasted pseudo measurements, and the virtual measurements are used to estimate the state of the active distribution network using the weighted least square (WLS) algorithm. Through simulations in the IEEE 33-bus system, it is shown that the proposed model provides accurate and reliable pseudo measurements for the active distribution network, improves the redundancy of the system, and thus further improves the accuracy of the state estimation and the capability of detecting and identifying bad data in active distribution systems without adding measurement devices.
关键词： Gaussian mixture model,Bad data,Forecasting of photovoltaic power generation,Active distribution system,State estimation,Pseudo measurement
更新于2025-09-16 10:30:52
-
Low Complexity Dimensioning of Sustainable Solar-enabled Systems: A Case of Base Station
摘要： Solar-enabled systems are becoming popular for provisioning pollution-free and cost-effective energy solution. Dimensioning of a solar-enabled system requires estimation of appropriate size of photovoltaic (PV) panel as well as storage capacity while satisfying a given energy outage constraint. Dimensioning has strong impact on the user’s quality of experience and network operator’s interest in terms of energy outage and revenue. In this paper, dimensioning problem of solar-enabled communication nodes is analyzed in order to reduce the computation overhead, where stand-alone solar-enabled base station (SS-BS) is considered as a case study. For this purpose, hourly solar data of last ten years has been taken into consideration for analysis. First, the power consumption model of BS is revised to save energy and increase revenue. Using the hourly solar data and power consumption profile, the lower bounds on panel size and storage capacity are obtained using Gaussian mixture model, which provides a reduced search space for cost-optimal system dimensioning. Then, the cost function and energy outage probability are modeled as functions of panel size and number of battery units using curve fitting technique. The cost function is proven to be quasiconvex, whereas energy outage probability is proven to be convex function of panel size and number of battery units. These properties transform the cost-optimal dimensioning problem into a convex optimization framework, which ensures a global optimal solution. Finally, a Computationally-efficient Energy outage aware Cost-optimal Dimensioning Algorithm (CECoDA) is proposed to estimate the system dimension without requiring exhaustive search. The proposed framework is tested and validated on solar data of several cities; for illustration purpose, four cities, New Delhi, Itanagar, Las Vegas, and Kansas, located at diverse geographical regions, are considered. It is demonstrated that, the presented optimization framework determines the system dimension accurately, while reducing the computational overhead up to 94% and the associated energy requirement for computation with respect to the exhaustive search method used in the existing approaches. The proposed framework CECoDA takes advantage of the location-dependent unique solar profile, thereby achieving cost-efficient solar-enabled system design in significantly less time.
关键词： computation efficiency,cost-optimal system dimensioning,Sustainable solar-enabled system,solar energy harvesting,energy outage,Gaussian mixture model,convex optimization,curve fitting
更新于2025-09-12 10:27:22
-
Modulation Mode Detection & Classification for in-Vivo Nano-Scale Communication Systems Operating in Terahertz Band
摘要： This work initiates the efforts to design an intelligent/cognitive nano receiver operating in Terahertz (THz) band. Specifically, we investigate two essential ingredients of an intelligent nano receiver—modulation mode detection (to differentiate between pulse based modulation and carrier based modulation), and modulation classification (to identify the exact modulation scheme in use). To implement modulation mode detection, we construct a binary hypothesis test in nano-receiver’s passband, and provide closed-form expressions for the two error probabilities. As for modulation classification, we aim to represent the received signal of interest by a Gaussian mixture model (GMM). This necessitates the explicit estimation of the THz channel impulse response, and its subsequent compensation (via deconvolution). We then learn the GMM parameters via Expectation-Maximization algorithm. We then do Gaussian approximation of each mixture density to compute symmetric Kullback-Leibler divergence in order to differentiate between various modulation schemes (i.e., M -ary phase shift keying, M -ary quadrature amplitude modulation). The simulation results on mode detection indicate that there exists a unique Pareto-optimal point (for both SNR and the decision threshold) where both error probabilities are minimized. The main takeaway message by the simulation results on modulation classification is that for a pre-specified probability of correct classification, higher SNR is required to correctly identify a higher order modulation scheme. On a broader note, this work should trigger the interest of the community in the design of intelligent/cognitive nano receivers (capable of performing various intelligent tasks, e.g., modulation prediction etc.).
关键词： nano-scale communication,Expectation-Maximization algorithm,Kullback-Leibler divergence,Gaussian mixture model (GMM),modulation classification,Terahertz (THz) band,modulation mode detection
更新于2025-09-09 09:28:46
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia, Spain (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Fast Sample Generation with Variational Bayesian for Limited Data Hyperspectral Image Classification
摘要： Labeling data for hyperspectral remote sensing image classification is a tedious and cost-intensive task. As a consequence, it is oftentimes necessary to perform classification when only very limited number of labeled training data is available. Several approaches have been proposed to address this problem. A recent proposal is to generate additional synthetic samples from a Gaussian Mixture Model for each class. One challenge with this approach lies in determining the number of components in the GMM. In this paper, we propose an approximation algorithm to select the number of components, namely Variational Bayesian (VB). The main advantage of VB is that it does not require multiple clustering computations in advance. Variational Bayesian not only greatly decreases the computational cost, but also generates comparable or better results in comparison to other methods.
关键词： synthetic data,hyperspectral remote sensing image classification,limited training data,Gaussian mixture model (GMM),Variational Bayesian (VB)
更新于2025-09-04 15:30:14