Traditional Matching Research Articles

A Feature Matching Method based on the Convolutional Neural Network

Nowadays, feature based 3D reconstruction and tracking technology have been widely used in the medical field. Feature matching is the most important step in feature-based 3D reconstruction process, as the accuracy of feature matching directly affects the accuracy of subsequent 3D point cloud coordinates. However, the matching performance of traditional feature matching methods is poor. To overcome this limitation, a method of matching based on convolutional neural network is presented. The convolutional neural network is trained by collecting a training set on the video sequence of a certain length from starting frame. The matched feature points in different endoscopic video frames are treated as the same category. The feature points in subsequent frames are matched by network classification. The proposed method is validated using the silicone simulation heart video and the endoscope video of the vivo beating heart obtained by Da Vinci’s surgical robot. Compared with SURF and ORB algorithms, as well as other methods, the experimental results show that the feature matching algorithm based on convolutional neural network is effective in the feature matching effect, rotation invariance, and scale invariance. For the first 200 frames of the video, the matching accuracy reached 90%.

Detection of crack defects in carbon fiber composites using passive infrared thermography

Aiming at the problem of low detection accuracy in detection of damage location, This article proposes a method to reduce the local jitter in passive infrared thermal imaging by using wavelet hierarchic hard threshold denoising based on template convolution, and thresholds are defined using an unsupervised Kmeans classification algorithm, the traditional SAD template convolution matching algorithm and the Sobel operator edge detection are improved. The denoizing algorithm achieves improved the denoising strength, the smoothness, and effectively extracts the crack defect features. To increase the effectiveness of nondestructive examination, the rapid detection method of temperature data matching is utilized before the infrared detection experiment, and the damage position was marked on the normal sample to allow for intuitive assessment of the damage location. The results show that by matching temperature data, the damage degree of the specimen can be quickly determined, and that the damage features can be accurately extracted using improved image processing algorithm.

A New Geomagnetic Vector Navigation Method Based on a Two-Stage Neural Network

The traditional geomagnetic matching navigation method is based on the correlation criteria operations between measurement sequences and a geomagnetic map. However, when the gradient of the geomagnetic field is small, there are multiple similar data in the geomagnetic database to the measurement value, which means the correlation-based matching method fails. Based on the idea of pattern recognition, this paper constructs a two-stage neural network by cascading a probabilistic neural network and a non-fully connected neural network to, respectively, classify geomagnetic vectors and their feature information in two steps: “coarse screening” and “fine screening”. The effectiveness and accuracy of the geomagnetic vector navigation algorithm based on the two-stage neural network are verified through simulation and experiments. In simulation, it is verified that when the geomagnetic average gradient is 5 nT/km, the traditional geomagnetic matching method fails, while the positioning accuracy based on the proposed method is 40.17 m, and the matching success rate also reaches 98.13%. Further, in flight experiments, under an average gradient of 11 nT/km, the positioning error based on the proposed method is 39.01 m, and the matching success rate also reaches 99.42%.

A physics‐based spectral matching (PBSM) method for generating fully site‐related ground motions

AbstractThe response spectrum is generally adopted in the seismic design to represent the seismic fortification level, but the structural dynamic analysis requires the ground motions as an input. However, ground motions generated by the traditional spectral matching methods do not have site‐related physical backgrounds for the target site. In this study, a physics‐based spectral matching (PBSM) method is developed to generate fully site‐related broadband ground motions (i.e., the generated ground motions have real physical backgrounds for the target site, including the source process, propagation path, and local site conditions) that are compatible to the target spectrum. In this method, the three‐dimensional (3D) numerical model around the target site is constructed to calculate the strain Green's tensors of all potential source locations using adjoint simulations. The variable space dimension of the seismic source is significantly reduced by applying the self‐similar feature of the multidimension source model, so that the optimization algorithm can be used to search for the rupture process that generates the physics‐based and spectrum‐matched ground motions (abbreviated to PBSM ground motions). The proposed method is applied to the Xiluodu dam in China. Compared with the traditional techniques, the generated ground motions have fully site‐related physical backgrounds and are compatible to the target spectrum. Additionally, as this method generates broadband ground motions based on the deterministic rupture process, any features of ground motions, such as large velocity pulses, can be taken into account throughout the optimization process. This study introduces the deterministic physical backgrounds of earthquakes to performance‐based seismic design and analysis. The proposed method may have a significant application potential in earthquake engineering.

Auto-tuning PVT data using multi-objective optimization: Application of NSGA-II algorithm

Reservoir simulation is known as perhaps the most widely used, accurate, and reliable method for field development in the petroleum industry. An integral part of a reliable reservoir simulation process is to consider robust and rigorous tuned EOS models. Traditionally, EOS models are tuned iteratively through arduous workflows against experimental PVT data. However, this comes with a number of drawbacks such as forcingly using weight factors, which upon alteration adversely affects the optimization process. The objective of the current work is thus to introduce an auto-tune PVT matching tool using NSGA-II multi-objective optimization. In order to illustrate the robustness of the presented technique, three different PVT samples are used, including two black-oil and one gas condensate sample. We utilize Peng-Robinson EOS during all the manual and auto-tuning processes. Comparison of auto-tuned EOS-generated results with those of experimental and computed statistical error values for these samples clearly show that the proposed method is robust. In addition, the proposed method, contrary to the manual matching process, provides the engineer with several matched solutions, which allows them to select a match based on the engineering background to be best amenable to the problem at hand. In addition, the proposed technique is fast, and can output several solutions within less time compared to the traditional manual matching method.

An INS/Geomagnetic Integrated Navigation Method for Coarse Estimation of Positioning Error and Search Area Adaption Applied to High-Speed Aircraft

The geomagnet aided inertial navigation algorithm has become one of the current research hotspots in the field of aircraft navigation. It can overcome the shortcomings of pure inertial navigation error accumulation over time and achieve long-term, autonomous, and all-weather navigation. The traditional iterative closest contour point (ICCP) algorithm, however, requires high sensor performance, has poor tolerance to noise, and has a complex algorithm with poor real-time performance. The timeliness of the traditional magnetic contour matching (MAGCOM) algorithm is related to the search area, the matching search area is fixed, and the algorithm is inflexible. To address the above problems, this article proposes a geomagnetic navigation method based on coarse estimation of navigation error and search area adaption. When the aircraft flies into the geomagnetic database region, the navigation error is roughly estimated based on the error characteristics of the inertial navigation system. Considering the influence of noise and uncertainty, three times the value of the estimated error is used as the edge length of the geomagnetic matching search area, and geomagnetic matching is performed for positioning. The product of velocity and time provided by the inertial navigation system is used as a constraint to determine whether the geomagnetic matching results are abnormal. If multiple consecutive geomagnetic matching results are normal, the navigation system automatically reduces the matching search area and search step to improve the matching accuracy and timeliness. Simulation experiments show that the accuracy of the adaptive search area geomagnetic matching method proposed in this article can reach within 200 m.

History Matching of Reservoir Simulation Model: a Case Study from the Mishrif Reservoir, Buzurgan Oilfield, Iraq

In petroleum reservoir engineering, history matching refers to the calibration process in which a reservoir simulation model is validated through matching simulation outputs with the measurement of observed data. A traditional history matching technique is performed manually by engineering in which the most uncertain observed parameters are changed until a satisfactory match is obtained between the generated model and historical information. This study focuses on step by step and trial and error history matching of the Mishrif reservoir to constrain the appropriate simulated model. Up to 1 January 2021, Buzurgan Oilfield, which has eighty-five producers and sixteen injectors and has been under production for 45 years when it started in 1976. Reservoir exhibits heterogeneity in porosity and permeability throughout the field, therefore it’s a big challenge to control all reservoir properties during matching process. The historical matching process includes matching field and wells oil and water production rates, water injection rates, water cut, and reservoir static pressure. Finally, the results show that the good matching between simulated model and observed data; oil and water production rates, water injection rates, water cut, and static reservoir pressure which allow for implementing perfect future production forecasting strategies.

Instability and Parametric Amplification of a Piezoelectric Energy Harvester Periodically Plucked by a Rotating Magnet

AbstractMagnetic plucking is an enabling technique to harvest energy from a rotary host as it converts the low-frequency excitation of rotational energy sources to high-frequency excitation that leads to resonance of small-scale piezoelectric energy harvesters. Traditional nonlinear analysis of the plucking phenomenon has relied on numerical integration methods. In this work, a semi-analytical method is developed to investigate the stability and bifurcation behaviors of rotary magnetic plucking, which integrates a second-order perturbation technique and discrete Fourier transform. Analysis through this method unfolds that the oscillatory response of the beam can lose stability through the saddle-node bifurcation and Hopf bifurcation, which eventually causes the beam to collide with the rotary host. Further, the influence of the magnetic gap and rotational speed on the stability is discussed. The study also reveals that the nonlinearity of the magnetic force can amplify the electrical power at primary resonance. As a result, the traditional impedance matching approach that neglects the nonlinearity of the magnetic force fails to predict the optimal electrical resistance. Finally, a finite element analysis shows that the instability is sensitive to damping, and the traditional single-mode approximation can lead to considerable error.

A glitch energy reduction method with low complexity dynamic element matching for data converters

Due to the appearance of huge glitch while using traditional Dynamic Element Matching (DEM) in data converters, different kinds of glitch-reduce structures are designed into DEM. However, most of these structures will work with high complexity. Based on random-rotation-binary-weighted selection (RRBS), this paper presents a novel low complexity glitch-reduced DEM algorithm, which can greatly reduce mismatch. The proposed DEM constructs glitch-reduce structure through data selector and comparator, so the complexity is reduced through using less components. As a result, the glitch-reduced DEM increases spurious free dynamic range (SFDR) to 92.47dB at 7.6MHz. Moreover, compared with conventional RRBS, glitch-reduced DEM makes the glitch reduce by 49% with low complexity in a 12-bit digital-to-analog converter (DAC).

5G ultra-dense network fingerprint positioning method based on matrix completion

The problem of high-precision indoor positioning in the 5G era has attracted more and more attention. A fingerprint location method based on matrix completion (MC-FPL) is proposed for 5G ultradense networks to overcome the high costs of traditional fingerprint database construction and matching algorithms. First, a partial fingerprint database constructed and the accelerated proximal gradient algorithm is used to fill the partial fingerprint database to construct a full fingerprint database. Second, a fingerprint database division method based on the strongest received signal strength indicator is proposed, which divides the original fingerprint database into several sub-fingerprint databases. Finally, a classification weighted K-nearest neighbor fingerprint matching algorithm is proposed. The estimated coordinates of the point to be located can be obtained by fingerprint matching in a sub-fingerprint database. The simulation results show that the MC-FPL algorithm can reduce the complexity of database construction and fingerprint matching and has higher positioning accuracy compared with the traditional fingerprint algorithm.

A new target localization method for image-guided radiation therapy of prostate cancer

In imaged-guided radiation therapy (IGRT), target localization is usually done with rigid-body registration based on anatomy matching. Problems arise when the target volume can only be matched partially due to inter-fractional organ motion and deformation, resulting in deteriorated target coverage and critical structure sparing. A new target localization method is investigated in which the treatment target volume is aligned with the prescription isodose surface. Our study included 15 prostate patients previously treated with intensity-modulated radiation therapy (IMRT). Patient setup and target localization were performed using a CT-on-rails system before and after the IMRT treatment. IMRT plans were generated on the original simulation CTs (15) and the same MUs and leaf sequences were used to compute the dose distributions on post-treatment CTs (98) with the isocenter adjustments based on either anatomical structure matching or prescription isodose surface alignment. When patients were aligned with the traditional anatomy matching method, the dose to 95% of the CTV, D95, received 74.0 – 77.6 Gy and the minimum CTV dose, Dmin, was 61.9 – 71.6 Gy, respectively, in the cumulative dose distributions. The rectal dose-volume constraints were violated in 35.7% of the treatment fractions. When patients were aligned using the new localization method, the dose to 95% of the CTV, D95, received 74.0 – 78.2 Gy and the minimum CTV dose, Dmin, was 68.4 – 71.6 Gy, respectively, in the cumulative dose distributions. The rectal dose-volume constraints were violated in 17.3% of the treatment fractions. Traditional IGRT target localization based on anatomy matching is effective for population-based PTV margins but not ideal for those patients with large inter-fractional prostate rotation/deformation due to large rectal and bladder volume variation. The new method using the prescription isodose surface to align the target volume could improve the target coverage and rectal sparing for these patients, which can be implemented clinically to improve target dose delivery accuracy.

Data-Driven Inversion-Free Workflow of Well Performance Forecast Under Uncertainty for Fractured Shale Gas Reservoirs

Abstract Well performance prediction and uncertainty quantification of fractured shale reservoir are crucial aspects of efficient development and economic management of unconventional oil and gas resources. The uncertainty related to the characterization of fracture topology is highly difficult to be quantified by the conventional model-based history matching procedure in practical applications. Data-space inversion (DSI) is a recently developed inversion-free and rapid forecast approach that directly samples the posterior distribution of quantities of interest using only prior model simulation results and historical data. This paper presents some comparative studies between a recent DSI implementation based on iterative ensemble smoother (DSI-IES), model-based history matching, and conventional decline curve analysis (DCA) for shale gas rate forecast. The DSI-IES method treats the shale gas production rate as target variables, which are directly predicted via conditioning to historical data. Dimensionality reduction is also used to regularize the time-series production data by low-order representation. This approach is tested on two examples with increasing complexity, e.g., a fractured vertical well and a multistage fractured horizontal well in the actual fractured Barnett shale reservoir. The results indicate that compared with the traditional history matching and DCA methods, the DSI-IES obtains high robustness with a high computational efficiency. The application of data-space inversion-free method can effectively tap the potential value directly from historical data, which provides theoretical guidance and technical support for rapid decision-making and risk assessment.

A Privacy-Preserving Ride Matching Scheme for Ride Sharing Services in a Hot Spot Area

Ride sharing is a service that enables users to share trips with others, conserving energy, decreasing emissions and reducing traffic congestion. Selecting a suitable partner for a user based on the their trip data is essential for the service, but it also leads to privacy disclosure, e.g., the user’s location and trajectory. Many privacy-preserving solutions for ride sharing services have been proposed, which are based on cryptographic technology and provide accurate matching services. However, these encryption-based algorithms are very complicated and difficult to calculate. In hot spots, such as stations, airports and sport gymnasiums, a large number of users may apply for a ride sharing service in short space of time, which will place huge pressure on the service provider. Using traditional matching methods increases the matching time and leads to a less favorable user experience. To solve these problems, we model them, aiming to maximize the vehicle’s carrying capacity and propose a lightweight privacy-preserving ride matching scheme for selecting feasible partners during busy periods with a large number of requests. To achieve this, we make use of the homomorphic encryption technique to hide location data and design a scheme to calculate the distances between users in road networks securely and efficiently. We employ a road network embedding technique to calculate the distance between users. Moreover, we use travel time instead of space distance, which makes matching more accurate. Further, with the encrypted itineraries of users, the service provider selects potential ride share partners according to the feasibility of time schedules. We use ciphertext packing to reduce overhead, improving the efficiency of ride matching. Finally, we evaluate our scheme with simulation and demonstrate that our scheme achieves an efficient and accurate matching service. It only takes a few seconds to complete the matching, and the matching accuracy is higher than 85 percent in most cases.

Water wave interaction with an array of submerged circular plates: Hankel transform approach

A Hankel transform based model is developed to study water wave interaction with an array of thin submerged horizontal perforated and/or elastic circular plates. The integral equations are formulated in terms of unknown functions related to the jump in velocity potential across each plate. A Galerkin method is adopted to the solution of these integral equations and the velocity potential jump across the plate is expressed in terms of Fourier-Gegenbauer series. The present model is found to be valid for multiple plates distributed arbitrarily, including the staggered arrangement, for which the traditional eigenfunction matching method would not work.

A Semantic Matching Method of E-Government Information Resources Knowledge Fusion Service Driven by User Decisions

This study focuses on the knowledge fusion model of e-government information resources that supports user decision-making information needs, it discusses the user decision-making information needs model, the knowledge fusion service model, and the relationship between them. The inter-layer mapping matching mechanism realizes the ultimate value of knowledge fusion. Therefore, this paper analyses and studies the mapping mechanism between the user information demand model and the knowledge fusion service model. A semantic, similarity-based knowledge fusion service matching method for e-government information resources is proposed to address the problem of lack of semantics in traditional web service matching methods. This method uses the ontology description language OWL-S to map information requirement documents of user decisions and knowledge fusion service function documents into an ontology tree structure. The authors then use this as the basis to calculate the concept similarity and relationship similarity measures, and the service matching based on semantic similarity can be realized.

An Improved Particle Filter Based on Gravity Measurement Feature in Gravity-Aided Inertial Navigation System

The existing gravity matching algorithms are affected by the initial position error of the inertial navigation system (INS), the gravity measurement error, and the similarity of the gravity background map. Aiming at the above problems, an improved particle filter based on the gravity measurement feature (IPFBGMF) is proposed in this article. In the IPFBGMF, both the value and change characteristic of gravity measurements are considered, and a novel position acquisition method based on the gravity measurement feature is proposed, which can reduce the influence of the initial position error of INS. In addition, a new concept called direction measurement using the heading angle of INS is proposed to optimize the weight of particles in the PF. The PF with direction measurement can reduce the influence of the gravity measurement error and the similarity of the gravity background map. Furthermore, the robustness of the improved PF with the precise position is proven. Finally, a navigation strategy is designed to apply the proposed algorithms. Simulations show that IPFBGMF has the highest positioning accuracy compared with the traditional gravity matching algorithms.

An optimal load indirect matching method without parameter identification and system efficiency optimization

In some wireless charging applications where the coil spacing varies in real time, such as UAV, electric boat and tram, etc., the traditional direct impedance matching method is difficult to identify the mutual inductance timely and accurately, thus affecting the efficiency optimization effect of the system. In this paper, an indirect impedance matching method without parameter identification is proposed, this method is based on the characteristic that the optimal voltage gain of the resonator is only related to its inherent parameters, and impedance matching can be achieved by controlling the voltage gain in real time. To further improve the efficiency of the system, a single-sided detuning design method is used to achieve soft switching of the inverter. Based on the optimal voltage gain expression derived by using both the indirect impedance matching method and the single-sided detuning design method, a compound control strategy for a series-series-compensated topology with dual-side power control is proposed to improve efficiency and stabilize the output voltage. A hardware prototype is built and a peak DC-to-DC efficiency with the optimal output resistance RL at about 28.9 Ω is 91.58%. When the output resistance RL is 100 Ω, the efficiency improved by 7% after using the proposed strategy.

DRFnet: Dynamic receptive field network for object detection and image recognition.

Biological experiments discovered that the receptive field of neurons in the primary visual cortex of an animal's visual system is dynamic and capable of being altered by the sensory context. However, in a typical convolution neural network (CNN), a unit's response only comes from a fixed receptive field, which is generally determined by the preset kernel size in each layer. In this work, we simulate the dynamic receptive field mechanism in the biological visual system (BVS) for application in object detection and image recognition. We proposed a Dynamic Receptive Field module (DRF), which can realize the global information-guided responses under the premise of a slight increase in parameters and computational cost. Specifically, we design a transformer-style DRF module, which defines the correlation coefficient between two feature points by their relative distance. For an input feature map, we first divide the relative distance corresponding to different receptive field regions between the target feature point and its surrounding feature points into N different discrete levels. Then, a vector containing N different weights is automatically learned from the dataset and assigned to each feature point, according to the calculated discrete level that this feature point belongs. In this way, we achieve a correlation matrix primarily measuring the relationship between the target feature point and its surrounding feature points. The DRF-processed responses of each feature point are computed by multiplying its corresponding correlation matrix with the input feature map, which computationally equals to accomplish a weighted sum of all feature points exploiting the global and long-range information as the weight. Finally, by superimposing the local responses calculated by a traditional convolution layer with DRF responses, our proposed approach can integrate the rich context information among neighbors and the long-range dependencies of background into the feature maps. With the proposed DRF module, we achieved significant performance improvement on four benchmark datasets for both tasks of object detection and image recognition. Furthermore, we also proposed a new matching strategy that can improve the detection results of small targets compared with the traditional IOU-max matching strategy.

Bin-scanning: Segmentation of X-ray CT volume of binned parts using Morse skeleton graph of distance transform

X-ray CT scanners, due to the transmissive nature of X-rays, have enabled the non-destructive evaluation of industrial products, even inside their bodies. In light of its effectiveness, this study introduces a new approach to accelerate the inspection of many mechanical parts with the same shape in a bin. The input to this problem is a volumetric image (i.e., CT volume) of many parts obtained by a single CT scan. We need to segment the parts in the volume to inspect each of them; however, random postures and dense contacts of the parts prohibit part segmentation using traditional template matching. To address this problem, we convert both the scanned volumetric images of the template and the binned parts to simpler graph structures and solve a subgraph matching problem to segment the parts. We perform a distance transform to convert the CT volume into a distance field. Then, we construct a graph based on Morse theory, in which graph nodes are located at the extremum points of the distance field. The experimental evaluation demonstrates that our fully automatic approach can detect target parts appropriately, even for a heap of 50 parts. Moreover, the overall computation can be performed in approximately 30 min for a large CT volume of approximately 2000×2000×1000 voxels.

Online education based on distributed multi-layer data processing technology

In order to promote the intelligent development of online education and optimize the status quo of online education, an online education method based on distributed multi-layer data processing technology is proposed. The method integrates the database information and builds the BTREE index by calculating the distributed database resource information, the initial information sequence, the main information set and the trade-off coefficient of information attributes. The bidirectional index method is used to extract the core data segment of integrated data and establish the time series table. The schedule is used to constrain the data twice, and the coincidence degree of the initial data transmission path is compared with the actual path parameters after evaluation, so as to realize the automatic matching of the data transmission path. The experimental results show that compared with the traditional transmission path matching method, the designed automatic matching method improves the matching accuracy of the complete data transmission path by 24% and the broken data transmission path by 19%, which has practical advantages. It is proved that this method can promote the precise personality teaching of online education, so as to achieve the effect of optimizing online education. It can also provide reference for the intelligent development of other online education.