This paper presents a vision system that provides a robust identification and localization of 2-D objects in industrial scenes. A symbolic image description based on the polygonal approximation of the object silhouett...
详细信息
This paper presents a vision system that provides a robust identification and localization of 2-D objects in industrial scenes. A symbolic image description based on the polygonal approximation of the object silhouette is extracted in video real time by the use of dedicated hardware. A two-stage matching algorithm is proposed. At the first stage hypotheses for assignments of image to model polygons are generated together with hypotheses for the object's pose. Corresponding continuous measures of similarity are derived from the turning functions of the curves. At the second stage compatible matches of polygons are collected by using a voting scheme in transformation space. Experimental results reveal the fault tolerance of the vision system with regard to noisy and broken image contours, as well as partial occlusion of objects. The robustness and easy adaptability of the vision system make it well suited for a wide variety of applications.
An advanced fuzzy logic control scheme has been proposed for a micro-computer based power system stabilizer to enhance the overall stability of power systems. The proposed control scheme utilizes the PID information o...
详细信息
An advanced fuzzy logic control scheme has been proposed for a micro-computer based power system stabilizer to enhance the overall stability of power systems. The proposed control scheme utilizes the PID information of the generator speed. The input signal to the stabilizer is the real power output of a study unit. Simulations show the effectiveness of the advanced fuzzy logic control scheme.
Financial fraud detection is an important task to ensure the security of financial system. Graph neural networks has shown good results in the field of financial fraud detection. However there are problems of insuffic...
Financial fraud detection is an important task to ensure the security of financial system. Graph neural networks has shown good results in the field of financial fraud detection. However there are problems of insufficient data mining and category imbalance in heterogeneous graphs of financial transaction networks. Therefore, this paper proposes Metapath Graph neural networks(Metapath-GNN), a graph neural network model based on metapath subgraph, for detecting financial frauds in complex transaction networks and hidden pattern states. Firstly, the subgraph is generated based on predefined metapath patterns by the metapath subgraph generation module. And the node selection is adjusted using the attention mechanism to improve the adaptability to the category imbalance data; then, an aggregation module is utilized to combine the subgraph and full graph information to generate more representative node embeddings. The effective information is fully exploited to enhance the detection performance of the model. Metapath-GNN is extensively evaluated on public datasets YelpChi, Amazon and Elliptic. In addition, for Elliptic, a real-world financial transaction dataset, the data labeling cost is reduced by a semi-supervised learning approach that makes full use of unlabeled data for training. The optimal performance is also achieved in the comparison experiments with the advanced methods. Such as F1-macro, Area Under the Receiver Operating Characteristic Curve(AUC) and Geometric Mean(GMean), by 11.33%, 1.26%, and 7.00% on YelpChi, 1.75%, 1.31% and 1.22% on Amazon, respectively. In Elliptic key indicator F1 improved by 6.78%. In T-Finance key metrics F1 improved by 1.28% and AUC by 3.54%.
A chemical bonding model for defects and defect precursors at Si—SiO2 interfaces is presented. Bonding geometries of neutral and charged Si‐, O‐, N‐, and H‐atoms are described in terms of the valence shell electr...
A chemical bonding model for defects and defect precursors at Si—SiO2 interfaces is presented. Bonding geometries of neutral and charged Si‐, O‐, N‐, and H‐atoms are described in terms of the valence shell electron pair repulsion model, thereby identifying a qualitative distinction between charged (a) Si‐atoms, and charged (b) O‐, N‐, or H‐atoms. Threefold coordinated Si‐ atoms are stable in three charge states (+, 0, and −), and can be active as trapping and/or recombination centers depending on their charge state. In contrast, there is a direct relationship between charge state and bonding coordination for O‐, N‐, and H‐atoms, and as such their roles in defect generation processes are qualitatively different. Reaction mechanisms based on these differences in local bonding are (i) discussed for the generation of defects comprised of threefold coordinated Si‐atoms, and/or positively charged H‐, O‐, and N‐atoms, with coordinations of two, three, and four, respectively, and (ii) compared with experiments.
Using a Gaussian weighting function for the receiver aperture, we obtain a closed-form representation for the receiver-aperture averaging effect for the intensity fluctuation of a beam wave in the turbulent atmosphere...
详细信息
Using a Gaussian weighting function for the receiver aperture, we obtain a closed-form representation for the receiver-aperture averaging effect for the intensity fluctuation of a beam wave in the turbulent atmosphere. It is shown that, unlike for the plane-wave case, the power scintillations do not always decrease when the receiver aperture is increased. The reasons are that (1) the intensity fluctuations on the axis for a coherent beam-wave source are smaller than these off the axis and (2) the averaging effect cannot show up when the total beam is within a coherent patch (i.e., the coherence length is larger than the beamwidth).
We report a new addressing mechanism for quasi-distributed absorption sensors based on the frequency modulated continuous wave (FMCW) method. The sensor units consist of open-path micro-optic cells constructed from GR...
详细信息
We report a new addressing mechanism for quasi-distributed absorption sensors based on the frequency modulated continuous wave (FMCW) method. The sensor units consist of open-path micro-optic cells constructed from GRIN lenses, each of differing lengths. Coherence addressing of the cells using FMCW is achieved by the interferometric mixing of two signals originating from each cell (from the glass/air interfaces). The time delay between the two reflections, along with the linear frequency ramp of the source, gives rise to beat frequencies in the mixed output which are different for each cell. The connecting fibre length between two successive sensor cells is chosen to be much greater than the coherence length of the source so that the reflections from different cells do not interfere. The interference patterns of all sensor cells add up at the detector whereby each individual sensing cell is identified by its power spectrum in the frequency domain. We show theoretically and experimentally how individual cells can be addressed and the measured signals obtained by suitable choice of cell length, proper modulation of the source and appropriate signal processing.
To calculate the propagation characteristics of a graded-index optical fibre with a parabolic core, such as a dual-shape core optical fibre having a parabolic-index centre core, one needs to solve a vector wave equati...
详细信息
To calculate the propagation characteristics of a graded-index optical fibre with a parabolic core, such as a dual-shape core optical fibre having a parabolic-index centre core, one needs to solve a vector wave equation in the inhomogeneous core region. The paper derives a coupled second-order differential equation with respect to the radial functions linking to the transverse electric fields through the refractive-index distribution, and solve it by the perturbation method. Analytical results show that the first-order perturbed solution plays a dominant role in the vector wave solution, especially for the fundamental mode. In fact, normalised frequencies and waveguide dispersions calculated using the first-order solution hold at least seven significant figures in the single-mode region. Comparison between cutoff frequencies of the conventional square-law optical fibre calculated here and those obtained by the numerical methods also guarantees the accuracy of the first-order solution. Using the first-order solution, the paper examines the waveguide dispersion and the mode field distribution of the dual-shape core optical fibre. Results show that the fibre with a parabolic-index centre core is one of the best dispersion-shifted optical fibres. The waveguide dispersion can be flattened over a relatively wide range of the waveguide parameters, and the confinement of modal power is good.
A micro-computer based fuzzy logic power system stabilizer is applied to a micro-machine system to investigate its efficiency in real time control. The stabilizing signal is determined by using measured speed or real ...
详细信息
A micro-computer based fuzzy logic power system stabilizer is applied to a micro-machine system to investigate its efficiency in real time control. The stabilizing signal is determined by using measured speed or real power signals at every sampling time to damp the system oscillations. The results show the proposed stabilizer improves the system damping effectively subject to various types of disturbances.
The winter season worldwide causes serious driving issues due to heavy fog on the roads. In situations like these, having an intelligent transportation system is crucial for safe driving. Therefore, to ensure safe and...
详细信息
In recent years, network intrusion refers to unauthorized access to systems or networks by external attackers or insiders aimed at stealing information or deploy malicious software. Traditional approaches for anomaly-...
详细信息
ISBN:
(数字)9798331508456
ISBN:
(纸本)9798331508463
In recent years, network intrusion refers to unauthorized access to systems or networks by external attackers or insiders aimed at stealing information or deploy malicious software. Traditional approaches for anomaly-based intrusion detection systems have faced several challenges which include high computational overhead and limited capability in detecting zero-day attacks due to reliance on predefined attack signatures or handcrafted features. To address these limitations, this research proposes a Machine Learning-based Stacking Ensemble Model (ML-SEM) for anomaly-based intrusion detection, effectively reducing computational overhead and enhancing detection of previously unseen attacks. Initially, data is collected from University of Nevada, Reno Intrusion Detection Dataset (UNR-IDD) which consists of network traffic data and various types of attacks. Then, the collected data is preprocessed by using standard scaler normalization which ensures that all features have a mean of zero and a standard deviation of one. After that, the features are selected using Principal Component Analysis (PCA) which reduces dimensionality and enhances model efficiency by eliminating redundant and less significant features. Finally, anomaly-based intrusion detection is done by using proposed ML-SEM. The proposed ML-SEM achieved better results in terms of accuracy (98.6%), precision (95.5%) and recall (98.3%) when compared with existing Random Forest (RF).
暂无评论