Search

Modern positron emission tomography (PET) detectors are made from pixelated scintillation crystal arrays and readout by Anger logic. The interaction position of the gamma-ray should be assigned to a crystal using a crystal position map or look-up table. Crystal identification is a critical procedure for pixelated PET systems. In this paper, we propose a novel crystal identification method for a dual-layer-offset LYSO based animal PET system via Lu-176 background radiation and mean shift algorithm. Single photon event data of the Lu-176 background radiation are acquired in list-mode for 3 h to generate a single photon flood map (SPFM). Coincidence events are obtained from the same data using time information to generate a coincidence flood map (CFM). The CFM is used to identify the peaks of the inner layer using the mean shift algorithm. The response of the inner layer is deducted from the SPFM by subtracting CFM. Then, the peaks of the outer layer are also identified using the mean shift algorithm. The automatically identified peaks are manually inspected by a graphical user interface program. Finally, a crystal position map is generated using a distance criterion based on these peaks. The proposed method is verified on the animal PET system with 48 detector blocks on a laptop with an Intel i7-5500U processor. The total runtime for whole system peak identification is 67.9 s. Results show that the automatic crystal identification has 99.98% and 99.09% accuracy for the peaks of the inner and outer layers of the whole system respectively. In conclusion, the proposed method is suitable for the dual-layer-offset lutetium based PET system to perform crystal identification instead of external radiation sources.

Background: The computer mouse is an important human-computer interaction device. But patients with physical finger disability are unable to operate this device. Surface EMG (sEMG) can be monitored by electrodes on the skin surface and is a reflection of the neuromuscular activities. Therefore, we can control limbs auxiliary equipment by utilizing sEMG classification in order to help the physically disabled patients to operate the mouse., Objective: To develop a new a method to extract sEMG generated by finger motion and apply novel features to classify sEMG., Methods: A window-based data acquisition method was presented to extract signal samples from sEMG electordes. Afterwards, a two-dimensional matrix image based feature extraction method, which differs from the classical methods based on time domain or frequency domain, was employed to transform signal samples to feature maps used for classification. In the experiments, sEMG data samples produced by the index and middle fingers at the click of a mouse button were separately acquired. Then, characteristics of the samples were analyzed to generate a feature map for each sample. Finally, the machine learning classification algorithms (SVM, KNN, RBF-NN) were employed to classify these feature maps on a GPU., Results: The study demonstrated that all classifiers can identify and classify sEMG samples effectively. In particular, the accuracy of the SVM classifier reached up to 100%., Conclusions: The signal separation method is a convenient, efficient and quick method, which can effectively extract the sEMG samples produced by fingers. In addition, unlike the classical methods, the new method enables to extract features by enlarging sample signals' energy appropriately. The classical machine learning classifiers all performed well by using these features.

This paper investigates the problem of locating a continuous chemical source using the concentration measurements provided by a wireless sensor network (WSN). Such a problem exists in various applications: eliminating explosives or drugs, detecting the leakage of noxious chemicals, etc. The limited power and bandwidth of WSNs have motivated collaborative in-network processing which is the focus of this paper. We propose a novel distributed least-squares estimation (DLSE) method to solve the chemical source localization (CSL) problem using a WSN. The DLSE method is realized by iteratively conducting convex combination of the locally estimated chemical source locations in a distributed manner. Performance assessments of our method are conducted using both simulations and real experiments. In the experiments, we propose a fitting method to identify both the release rate and the eddy diffusivity. The results show that the proposed DLSE method can overcome the negative interference of local minima and saddle points of the objective function, which would hinder the convergence of local search methods, especially in the case of locating a remote chemical source.

This paper addresses the collective odor source localization (OSL) problem in a time-varying airflow environment using mobile robots. A novel OSL methodology which combines odor-source probability estimation and multiple robots' search is proposed. The estimation phase consists of two steps: firstly, the separate probability-distribution map of odor source is estimated via Bayesian rules and fuzzy inference based on a single robot's detection events; secondly, the separate maps estimated by different robots at different times are fused into a combined map by way of distance based superposition. The multi-robot search behaviors are coordinated via a particle swarm optimization algorithm, where the estimated odor-source probability distribution is used to express the fitness functions. In the process of OSL, the estimation phase provides the prior knowledge for the searching while the searching verifies the estimation results, and both phases are implemented iteratively. The results of simulations for large-scale advection-diffusion plume environments and experiments using real robots in an indoor airflow environment validate the feasibility and robustness of the proposed OSL method.

This article presents a hysteresis modeling algorithm to describe the asymmetric nonlinearity in piezoelectric actuators (PEAs). The proposed algorithm primarily consists of two parts: 1) the wiping-out mechanism, and 2) the description method for the current hysteresis curve (CHC). The first part is associated with the realization of the wiping-out mechanism. It is used to determine the hysteresis curve where the hysteresis trajectory moves along, and this curve represents the CHC. In the second part, the information of the major ascending or descending curve is used to describe the CHC, and meanwhile, the trajectory is also described. The proposed algorithm works in real-time, which provides an intuitive mechanism to understand and investigate the asymmetric hysteresis nonlinearity in PEAs. Furthermore, the analytical inverse modeling algorithm is proposed to reduce the hysteresis effect. Experiments are conducted to validate the effectiveness of the proposed algorithms. [ABSTRACT FROM AUTHOR]

The key to successful detection of a localized bearing fault lies in extracting fault-induced impulse responses. Unfortunately, these responses are often contaminated by background noise. The popular K-singular value decomposition (K-SVD) denoising algorithm can be used to extract impulse responses from noise, but it may obtain weak impulse responses against reliable fault detection. An SOSO boosting technique is proposed to improve the performance of the K-SVD denoising algorithm. Given an initial denoised signal, this paper proposes iteratively repeating the following SOSO boosting procedure: 1) Strengthen the underlying signal by adding the previous denoised signal to the noisy input signal; 2) Operate the K-SVD denoising algorithm on the strengthened signal; 3) Subtract the previous denoised signal from the restored signal-strengthened outcome; and 4) Operate the modified K-SVD denoising algorithm again on the resulting signal after the subtraction. The last step, i.e., the secondary denoising step, plays a crucial role in preventing the enhancement of the residual noise that originates from the first denoised result. The results of numerical and experimental studies show that the SOSO-based K-SVD denoising algorithm not only enhances weak impulse responses significantly but also reduces potential residual noise effectively compared to the SOS-based and the original K-SVD denoising algorithms. [ABSTRACT FROM AUTHOR]

In this paper, we investigate the energy-efficient resource allocation problem in an uplink non-orthogonal multiple access (NOMA) millimeter wave system, where the fully-connected-based sparse radio frequency chain antenna structure is applied at the base station (BS). To relieve the pilot overhead for channel estimation, we propose a codebook-based analog beam design scheme, which only requires to obtain the equivalent channel gain. On this basis, users belonging to the same analog beam are served via NOMA. Meanwhile, an advanced NOMA decoding scheme is proposed by exploiting the global information available at the BS. Under predefined minimum rate and maximum transmit power constraints for each user, we formulate a max-min user energy efficiency (EE) optimization problem by jointly optimizing the detection matrix at the BS and transmit power at the users. We first transform the original fractional objective function into a subtractive one. Then, we propose a two-loop iterative algorithm to solve the reformulated problem. Specifically, the inner loop updates the detection matrix and transmit power iteratively, while the outer loop adopts the bi-section method. Meanwhile, to decrease the complexity of the inner loop, we propose a zero-forcing (ZF)-based iterative algorithm, where the detection matrix is designed via the ZF technique. Finally, simulation results show that the proposed schemes obtain a better performance in terms of spectral efficiency and EE than the conventional schemes. [ABSTRACT FROM AUTHOR]

Originating from a nearest point problem, a novel method called one-class classification based on the convex hull (OCCCH) is proposed for one-class classification problems. The basic goal of OCCCH is to find the nearest point to the origin from the reduced convex hull of training samples. A generalized Gilbert algorithm is proposed to solve the nearest point problem. It is a geometric algorithm with high computational efficiency. OCCCH has two different forms, i.e., OCCCH-1 and OCCCH-2. The relationships among OCCCH-1, OCCCH-2 and one-class support vector machine (OCSVM) are investigated theoretically. The classification accuracy and the computational efficiency of the three methods are compared through the experiments conducted on several benchmark datasets. Experimental results show that OCCCH (including OCCCH-1 and OCCCH-2) using the generalized Gilbert algorithm performs more efficiently than OCSVM using the well-known sequential minimal optimization (SMO) algorithm; at the same time, OCCCH-2 can always obtain comparable classification accuracies to OCSVM. Finally, these methods are applied to the monitoring model constructions for bearing fault detection. Compared with OCCCH-2 and OCSVM, OCCCH-1 can significantly decrease the false alarm ratio while detecting the bearing fault successfully. [ABSTRACT FROM AUTHOR]

Cosmic ray muon tomography is a novel technology to detect high-Z material. A prototype of TUMUTY with 73.6 cm×73.6 cm large scale position sensitive MRPC detectors has been developed and is introduced in this paper. Three test kits have been tested and image is reconstructed using MAP algorithm. The reconstruction results show that the prototype is working well and the objects with complex structure and small size (20 mm) can be imaged on it, while the high-Z material is distinguishable from the low-Z one. This prototype provides a good platform for our further studies of the physical characteristics and the performances of cosmic ray muon tomography. [ABSTRACT FROM AUTHOR]

This paper describes an easy-to-use system to estimate the shape of a human body and his/her clothes. The system uses a Kinect to capture the human׳s RGB and depth information from different views. Using the depth data, a non-rigid deformation method is devised to compensate motions between different views, thus to align and complete the dressed shape. Given the reconstructed dressed shape, the skin regions are recognized by a skin classifier from the RGB images, and these skin regions are taken as a tight constraints for the body estimation. Subsequently, the body shape is estimated from the skin regions of the dressed shape by leveraging a statistical model of human body. After the body estimation, the body shape is non-rigidly deformed to fit the dressed shape, so as to extract the cloth field of the dressed shape. We demonstrate our system and the therein algorithms by several experiments. The results show the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

This paper discusses odor source localization (OSL) using a mobile robot in an outdoor time-variant airflow environment. A novel OSL algorithm based on particle filters (PF) is proposed. When the odor plume clue is found, the robot performs an exploratory behavior, such as a plume-tracing strategy, to collect more information about the previously unknown odor source. In parallel, the information collected by the robot is exploited by the PF-based OSL algorithm to estimate the location of the odor source in real time. The process of the OSL is terminated if the estimated source locations converge within a given small area. The Bayesian-inference-based method is also performed for comparison. Experimental results indicate that the proposed PF-based OSL algorithm performs better than the Bayesian-inference-based OSL method. [ABSTRACT FROM AUTHOR]

In order to eliminate the influence of periodic disturbances on the rate smoothness of the system, an adapting algorithm based on the internal model principle is studied. This method can eliminate periodic disturbances caused by single sinusoidal signal, multi-sinusoidal signal, and even time-varying sinusoidal signal. Furthermore, the ripple phenomenon occurred in the frequency estimation is studied when the frequency of disturbances changes rapidly. The method is simulated in test turn table, and the result shows that the method has the capability to identify the frequency of periodic disturbances and restrain periodic disturbances with uncertain frequency. [ABSTRACT FROM AUTHOR]

3D shapes and 2D images usually contain complementary information for each other, and thus joint analysis of both of them will benefit some problems existing in different domains. Leveraging the connection between 2D images and 3D shapes, it's potential to mine lacking information of one modal from the other. Stemming from this insight, we design and implement a CNN architecture to jointly analyze shapes and images even with few training data guidance. The core of our architecture is a domain adaptation algorithm, which builds up the connection between underlying feature spaces of images and shapes, then aligns and correlates the intrinsic structures therein. The proposed method facilitates the recognition and retrieval tasks. Experiments on the shape recognition tasks show that our approach has superior performance under the difficult setting: zero-shot learning and few-shot learning. We also evaluate our method on the retrieval tasks, and demonstrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Locally linear embedding (LLE) is a classical non-linear dimension reduction algorithm. LLE can deal with a nonlinear dataset and extract the significant feature from the dataset. However, LLE may fail, when signal is non-stationary or contains strong noise. In this paper, wavelet transformation, singular value decomposition (SVD), and LLE are assembled to extract the feature of a dataset based on which a new fault diagnosis method is developed for submersible plunger pump fault recognition. First, each sample is preprocessed by wavelet transformation and SVD, upon which a new feature space is constructed; then, LLE is utilized to reduce the dimensions of the feature space; finally, support vector machines (SVM) is invoked to recognize the device status. The experiments performed on submersible plunger pump dataset and bearing dataset demonstrate that the proposed method is effective and can achieve a high diagnosis accuracy. [ABSTRACT FROM AUTHOR]

Video segmentation is an important preprocessing step for many computer vision and graphics tasks. Its main goal is to group the voxels in the video volume with similar appearance and motion into spatio-temporally consistent supervoxels. In this paper, we formulate video segmentation as an L 0 gradient minimization problem, so that the spatio-temporal coherence can be effectively enforced through a gradient sparsity pursuit way. In our method, the appearance and motion descriptor space is first built for over-segmented image patches of each video frame. Then the L 0 gradient minimization is performed in the descriptor space, for both spatial and temporal dimensions. To solve the non-convex L 0 norm minimization problem, we extend the fused coordinate descent algorithm from 2D image grids to 3D video volume. We conduct quantitative evaluation of our method in a public video segmentation benchmark LIBSVX. The experimental results demonstrate our superior performance to state-of-the-arts in segmentation accuracy and undersegmentation error, and comparable performance in boundary recall and explained variation. [ABSTRACT FROM AUTHOR]

Local linear embedding (LLE) algorithm is widely utilized to feature extraction for fault diagnosis, but the diagnosis result is sensitive to reconstruction weight W of LLE. To make W more significant and robust, in this paper, ISLLE algorithm is proposed with the aid of iterative shrinkage technology and LLE algorithm. In ISLLE algorithm, a surrogate function is introduced, upon which the high-dimensional optimization problem can be decoupled into a set of one-dimensional equations, then W can be easily computed by iterative shrinkage method. In each iteration, the small and negative weight coefficients are eliminated, while the large ones are shrunk, which can be regarded as feature extraction and noise reduction. Hence, the signals processed by ISLLE are more beneficial to diagnosis. Three real datasets are used to examine the proposed method. The experimental results demonstrate that the proposed method is valid, and the performance of ISLLE outperforms that of the original LLE. [ABSTRACT FROM AUTHOR]

Abstract This paper presents a two-stage optimization method for a coupled capacity planning and operation problem, cast within the economical operation of Regional Integrated Energy System. The first stage optimization of the proposed model represents a regional integrated energy system planner whose purpose is to minimize its energy and environmental cost, while the second stage is an operation problem whose primary role is to achieve the optimal operation scheme of the system. The regional integrated energy system planner pursues best interests by co-optimizing the capacity configuration and power output of individual energy supply module, while the regional integrated energy system maximizes the installed capacity of renewable energy sources and minimizes the environmental costs. To illustrate the advantage of the proposed method, the NSGA-II algorithm and the mixed integer linear programming method are implemented to solve the model based on simulation. Besides, application of the optimization method proposed to the energy infrastructure of a regional integrated energy system in China is discussed, and the results obtained through simulation are compared to the bi-level optimization objectives. The results show that the proposed method is economical and effective in practical application. Highlights • A bi-level optimization model for integrated energy system is presented. • A decentralized optimization method is proposed for solving capacity planning. • Accuracy and economy of capacity planning are improved. • The economic efficiency of the RIES is improved. [ABSTRACT FROM AUTHOR]

A new framework for image retrieval/object search is proposed based on the VLAD model and SURF descriptors to improve the codebook construction speed, the image matching accuracy, and the online retrieval speed and to reduce the data storage. First, SURF binarization and dimensionality reduction methods are proposed to convert a 64-dimensional SURF descriptor into an 8-dimensional descriptor. Second, a two-step clustering algorithm is proposed for codebook construction to significantly reduce the computational cost of clustering while maintaining the accuracy of the clustering results. Moreover, for object search, a scalable overlapping partition method is proposed to segment an image into 65 patches with different sizes so that the object can be matched quickly and efficiently. Finally, a feature fusion strategy is employed to compensate the performance degradation caused by the information loss of our proposed dimensionality reduction method. Experiments on the Holidays and Oxford datasets demonstrate the effectiveness and efficiency of the proposed algorithms. [ABSTRACT FROM AUTHOR]

As an integral feature of a future smart grid, demand response (DR) provides utility companies an emerging alternative to boost the reliability performance (e.g., supply adequacy) of power systems. Unlike for physical devices, the availability of DR is not only dependent on the operations of electric appliances on the demand side but can be affected by customers’ behaviors. Thus, how or to what extent DR actions can affect power system reliability becomes an important issue for utilities. In this paper, a new approach for assessing the contribution of incentive-based DR to the supply adequacy of smart distribution systems (SDS) is presented. Compared with existing methods, our method explicitly captures the varying availability of customer DR capabilities. We introduce a behavior-reinforcement procedure to model the correlation of users’ participation willingness with their historical DR profitability. Then, a strategic DR dispatch strategy can be developed to optimize users’ DR availability bids in the market. In addition, the interdependency between a communication system and DR operation has also been considered. A hybrid algorithm based on sequential Monte-Carlo simulation and an optimal load dispatch method is employed to evaluate the system reliability in this context. The proposed approach is illustrated using both a small-scale test case and a real regional distribution grid in China. The results demonstrate the effectiveness of the presented method. [ABSTRACT FROM AUTHOR]

The Tsinghua University MUon Tomography facilitY (TUMUTY) has been built up and it is utilized to reconstruct the special objects with complex structure. Since fine image is required, the conventional Maximum likelihood Scattering and Displacement (MLSD) algorithm is employed. However, due to the statistical characteristics of muon tomography and the data incompleteness, the reconstruction is always instable and accompanied with severe noise. In this paper, we proposed a Maximum a Posterior (MAP) algorithm for muon tomography regularization, where an edge-preserving prior on the scattering density image is introduced to the object function. The prior takes the l p norm ( p >0) of the image gradient magnitude, where p =1 and p =2 are the well-known total-variation (TV) and Gaussian prior respectively. The optimization transfer principle is utilized to minimize the object function in a unified framework. At each iteration the problem is transferred to solving a cubic equation through paraboloidal surrogating. To validate the method, the French Test Object (FTO) is imaged by both numerical simulation and TUMUTY. The proposed algorithm is used for the reconstruction where different norms are detailedly studied, including l 2 , l 1 , l 0.5 , and an l 2–0.5 mixture norm. Compared with MLSD method, MAP achieves better image quality in both structure preservation and noise reduction. Furthermore, compared with the previous work where one dimensional image was acquired, we achieve the relatively clear three dimensional images of FTO, where the inner air hole and the tungsten shell is visible. [ABSTRACT FROM AUTHOR]

The hybrid pixel detector Timepix3 allows the measurement of the time and energy deposition of an event simultaneously in each 55 μ m pixel, which makes Timepix3 a promising approach for a compact Compton camera. However, the angular resolution of Compton camera based on this kind of detector with high pixel density is usually degraded in imaging of MeV gamma-ray sources, because the diffusion of energetic Compton electron or photoelectron could trigger many pixels and lead to an inaccurate measurement of interaction position. In this study, an electron track algorithm is used to reconstruct the electron track and determine the interaction point. An demonstrative experiment was carried out, showing that the effect of this algorithm was significant. The angular resolution measures of a single layer Compton camera based on Timepix3 was enhanced to 12 degrees (FWHM) in imaging of a 60Co gamma-ray source. [ABSTRACT FROM AUTHOR]