Probabilistic Fusion Research Articles

UAV clusters information geometry fusion positioning method with LEO satellite system

The existing Low-Earth-Orbit (LEO) positioning performance cannot meet the requirements of Unmanned Aerial Vehicle (UAV) clusters for high-precision real-time positioning in the Global Navigation Satellite System (GNSS) denial conditions. Therefore, this paper proposes a UAV Clusters Information Geometry Fusion Positioning (UC-IGFP) method using pseudo-ranges from the LEO satellites. A novel graph model for linking and computing between the UAV clusters and LEO satellites was established. By utilizing probability to describe the positional states of UAVs and sensor errors, the distributed multivariate Probability Fusion Cooperative Positioning (PF-CP) algorithm is proposed to achieve high-precision cooperative positioning and integration of the cluster. Criteria to select the centroid of the cluster were set. A new Kalman filter algorithm that is suitable for UAV clusters was designed based on the global benchmark and Riemann information geometry theory, which overcomes the discontinuity problem caused by the change of cluster centroids. Finally, the UC-IGFP method achieves the LEO continuous high-precision positioning of UAV clusters. The proposed method effectively addresses the positioning challenges caused by the strong direction of signal beams from LEO satellites and the insufficient constraint quantity of information sources at the edge nodes of the cluster. It significantly improves the accuracy and reliability of LEO-UAV cluster positioning. The results of comprehensive simulation experiments show that the proposed method has a 30.5% improvement in performance over the mainstream positioning methods, with a positioning error of 14.267 m.

Application of multiple spectrum domain fault probability fusion model in high-speed rail bearings

Application of multiple spectrum domain fault probability fusion model in high-speed rail bearings

Uncertainties quantification for damage localization in concrete based on Bayesian method

The presence of defects in concrete can diminish load-bearing capacity of structures, giving rise to potential concerns regarding safety and durability. Thus, a method that enhances the sensitivity, resolution and robustness of damage localization is critically necessary to assess the condition of concrete structures. This research presents a damage localization method based on Bayesian probabilistic fusion, and uncertainties from measurement and identification process are considered and quantified. The likelihood function is constructed based on the hyperbola-based damage localization method, and the posterior distributions of unknown parameters are calculated via Bayesian theorem combined with measurement data. Furthermore, a meso-level finite element model is established, wherein the concrete medium is considered as a three-phase composite material consisting of polygonal aggregates, mortar matrix and interface transition zones. Owing to the meso-level modeling, the propagation behavior of stress waves within concrete and complicated interactions between stress waves and concrete internal structures can be better characterized. Finally, the damage information, time-difference-of arrival, is extracted from the response signals and the efficiency of the proposed method is verified numerically. The numerical results demonstrate that the proposed probabilistic fusion method outperforms the conventional hyperbola-based method in terms of achieving high spatial resolution and resilience in damage localization.

A Robust Monocular and Binocular Visual Ranging Fusion Method Based on an Adaptive UKF.

Visual ranging technology holds great promise in various fields such as unmanned driving and robot navigation. However, complex dynamic environments pose significant challenges to its accuracy and robustness. Existing monocular visual ranging methods are susceptible to scale uncertainty, while binocular visual ranging is sensitive to changes in lighting and texture. To overcome the limitations of single visual ranging, this paper proposes a fusion method for monocular and binocular visual ranging based on an adaptive Unscented Kalman Filter (AUKF). The proposed method first utilizes a monocular camera to estimate the initial distance based on the pixel size, and then employs the triangulation principle with a binocular camera to obtain accurate depth. Building upon this foundation, a probabilistic fusion framework is constructed to dynamically fuse monocular and binocular ranging using the AUKF. The AUKF employs nonlinear recursive filtering to estimate the optimal distance and its uncertainty, and introduces an adaptive noise-adjustment mechanism to dynamically update the observation noise based on fusion residuals, thus suppressing outlier interference. Additionally, an adaptive fusion strategy based on depth hypothesis propagation is designed to autonomously adjust the noise prior of the AUKF by combining current environmental features and historical measurement information, further enhancing the algorithm's adaptability to complex scenes. To validate the effectiveness of the proposed method, comprehensive evaluations were conducted on large-scale public datasets such as KITTI and complex scene data collected in real-world scenarios. The quantitative results demonstrate that the fusion method significantly improves the overall accuracy and stability of visual ranging, reducing the average relative error within an 8 m range by 43.1% and 40.9% compared to monocular and binocular ranging, respectively. Compared to traditional methods, the proposed method significantly enhances ranging accuracy and exhibits stronger robustness against factors such as lighting changes and dynamic targets. The sensitivity analysis further confirmed the effectiveness of the AUKF framework and adaptive noise strategy. In summary, the proposed fusion method effectively combines the advantages of monocular and binocular vision, significantly expanding the application range of visual ranging technology in intelligent driving, robotics, and other fields while ensuring accuracy, robustness, and real-time performance.

Enhancing hydrological data completeness: A performance evaluation of various machine learning techniques using probabilistic fusion imputer with neural networks for streamflow data reconstruction

The present-day accessibility of streamflow data, particularly in the developing countries, is often marked by a multitude of data shortfalls or distortions. This study investigates the estimate of missing streamflow data using machine learning approaches, including K-nearestneighbour (KNN), Predictive Mean Matching (PMM), Random Forest (RF) and a novel technique of Probabilistic Fusion Imputer with Neural Networks (PROFINN). This study tackles the issue of data insufficiency in such time series considering the inclusion of numerous hydrological parameters by means of RF feature selector, to determine the most significant ones among them. The study explores the efficacy of selected models on various data gaps under different hydrological scenarios, including diverse flow characteristics (mean, high and low flows) and gap lengths (long and short gaps, continuous and discontinuous gaps) and presents a pattern of ranking system that assesses the level of suitability of each technique for various data gaps. This study underscore PROFINN’s remarkable performance across all scenarios, yielding an average Root Mean Square Error (RMSE) of 0.91 and an average Nash-Sutcliffe Efficiency (NSE) value of 0.93 when applied for an intermittent river system of Pamba in Southern Kerala, India. RF follows PROFINN in the imputation of extreme flows as well as long and short gap scenarios. KNN closely follows PROFINN for the imputation of continuous and discontinuous gap scenarios. This study augments the significance of tailored machine learning techniques in enhancing the integrity of hydrological datasets, offering valuable insights for effective decision-making in water resource management and related fields. Furthermore, the success of PROFINN in streamflow data imputation suggests its potential extension to other hydrological research domains, like historical data reconstruction assisting climate change studies and future water resources planning, emphasizing the broader relevance and applicability of this study’s findings.

A machine vision approach with temporal fusion strategy for concrete vibration quality monitoring

In concrete construction, ensuring the quality of vibration is paramount for maintaining the strength, durability, and quality of structures. This study proposed a method for monitoring the vibration quality of vibrating robots to replace subjective human judgment. The method employed an improved EfficientNetV2 model to classify the vibration quality into four levels: unqualified, middle, qualified, and over-vibration. A temporal fusion strategy was introduced, employing time-domain probability fusion to enhance the accuracy and stability of the results. Additionally, image patching was applied to reduce computational complexity while preserving feature integrity. Experimental results demonstrate that the proposed method outperforms common mainstream models, achieving an accuracy of 96.47%, with a relatively small parameter size of 13.8 M. Compared to non-temporal fusion strategy, the accuracy is improved by 2.06%. This research has been successfully applied in practical engineering, providing a reliable means of quality assurance for concrete structures and demonstrating potential application prospects in the field of engineering construction.

Research on Three-Dimensional Cloud Structure Retrieval and Fusion Technology for the MODIS Instrument

Accurate three-dimensional (3D) cloud structure measurements are critical for assessing the influence of clouds on the Earth’s atmospheric system. This study extended the MODIS (Moderate-Resolution Imaging Spectroradiometer) cloud vertical profile (64 × 64 scene, about 70 km in width × 15 km in height) retrieval technique based on conditional generative adversarial networks (CGAN) to construct seamless 3D cloud fields for the MODIS granules. Firstly, the accuracy and spatial continuity of the retrievals (of 7180 samples from the validation set) were statistically evaluated. Then, according to the characteristics of the retrieval error, a spatially overlapping-scene ensemble generation method and a bidirectional ensemble binning probability fusion (CGAN-BEBPF) technique were developed, which improved the CGAN retrieval accuracy and support to construct seamless 3D clouds for the MODIS granules. The CGAN-BEBPF technique involved three steps: cloud masking, intensity scaling, and optimal value selection. It ensured adequate coverage of the low reflectivity areas while preserving the high-reflectivity cloud cores. The technique was applied to retrieve the 3D cloud fields of Typhoon Chaba and a multi-cell convective system and the results were compared with ground-based radar measurements. The cloud structures of the CGAN-BEBPF results were highly consistent with the ground-based radar observations. The CGAN-EBEPF technique retrieved weak ice clouds at the top levels that were missed by ground-based radars and filled the gaps of the ground-based radars in the lower levels. The CGAN-BEBPF was automated to retrieve 3D cloud radar reflectivity along the MODIS track over the seas to the east and south of mainland China, providing valuable cloud information to support maritime and near-shore typhoons and convection prediction for the cloud-sensitive applications in the regions.

New approach for satellite DEM accuracy enhancement by combing machine learning, fuzzy majority voting, and weighted interpolation techniques

The digital elevation model (DEM) is crucial in many global and regional scientific studies in civilian and military applications. The aim of this research is to develop and test a new DEM approach for correcting the various errors in the Shuttle Radar Topography Mission (SRTM) digital elevation model. Firstly, the DEMs with the feature attributes from Sentinel-2 multispectral imagery are generated. Secondly, SRTM DEM with one band and attributes of a sentinel-2 image with eight bands are used as input data in supervised max-like hood, an artificial neural network (ANN), and support vector machine (SVM) classification models. Thirdly, ANN, supervised max-like hood, and SVM classification models, which have various properties, are fused by fuzzy majority voting (probability fusion). Finally, the fused probability is assigned for each pixel of the image, which has 12 fixed ground control points (GCPs), which is considered new input data for the inverse probability weighted interpolation (IPWI) approach to create the corrected SRTM elevations. The results were contrasted with a reference DEM (RD) created by image matching with Worldview-1 stereo satellite images, which had a 1-m vertical accuracy. The results of this study demonstrated that the RMSE of the original SRTM DEM was 5.95. On the other hand, the RMSE of the estimated elevations by the IPWI approach has been improved to 1.98 compared with that of the MLR method (3.01). The study shows a series of significant improvements in the SRTM when assessed with the reference DEM, with an RMSE reduction of (66.72%) when compared to the widely utilized multiple linear regression (MLR) method. It can be concluded that the elevation error of the original SRTM DEM is clearly reduced by the suggested approach.

Fall Recognition Based on Time-Level Decision Fusion Classification

We propose a vision-based fall detection algorithm using advanced deep learning models and fusion methods for smart safety management systems. By detecting falls through visual cues, it is possible to leverage existing surveillance cameras, thus minimizing the need for extensive additional equipment. Consequently, we developed a cost-effective fall detection system. The proposed system consists of four modules: object detection, pose estimation, action recognition, and result fusion. Constructing the fall detection system involved the utilization of state-of-the-art (SOTA) models. In the fusion module, we experimented with various approaches, including voting, maximum, averaging, and probabilistic fusion. Notably, we observed a significant performance improvement with the use of probabilistic fusion. We employed the HAR-UP dataset to demonstrate this enhancement, achieving an average 0.84% increase in accuracy compared to the baseline, which did not incorporate fusion methods. By applying our proposed time-level ensemble and skeleton-based fall detection approach, coupled with the use of enhanced object detection and pose estimation modules, we substantially improved the robustness and accuracy of the system, particularly for fall detection in challenging scenarios.

Fracture-vuggy carbonate reservoir characterization based on multiple geological information fusion

The complexity and strong heterogeneity of carbonate reservoirs with fracture-vuggy structures present significant challenges in reservoir characterization. To address these challenges, we propose a novel multi-element information fusion modeling approach. This approach is designed to integrate multiple methods and incorporate multi-probability fusion at various facies and scales, thereby bridging the gap between geological information and reservoir modeling. Our methodology involves four key steps. First, the statistics between frequency of karst and geological information are acquired, and we quantify the statistics to regression equations. Second, these regression equations are transferred to probability bodies. The probability bodies can be applied in modeling as a soft control. But just one single body can be input in modeling process. Third, multiple probability bodies are fused into a fusion probability body by a probability fusion algorithm, which can keep the potential information of probability bodies. Finally, we apply the probability body in modeling workflow. By this way, the fusion method bridges the gap between geological information and modeling. The model established through our proposed method showed a significant level of consistency with reservoir re-evaluation, achieving an impressive 90% degree of alignment. Furthermore, the history match analysis revealed a high correlation, indicating the model's reliability. The method effectively integrates various scales and types of geological information, offering an accurate approach to complex carbonate reservoir modeling.

BERT based Blended approach for Fake News Detection

This paper presents a new approach for detecting fake news on social media. Previous works in this domain have demonstrated that context is an important factor when attempting to distinguish subtle differences within text. Fake news itself presents different level of difficulty due the vast similarity that exists between genuine and fake news contents. Therefore, we propose a collaborative approach which uses probabilistic fusion strategy to combine the knowledge gained from modelling two language models, BERT-LSTM and BERT-CNN. To achieve the fusion, we exploit the Bayesian method. Our experiments are conducted on two fake news detection datasets. The detection accuracy attained in these experiments attest to the efficiency of the proposed method, as our approach is very competitive compared to the state-of-the-art methods.

Damage Location Method of Pipeline Structure by Ultrasonic Guided Wave Based on Probability Fusion

Damage Location Method of Pipeline Structure by Ultrasonic Guided Wave Based on Probability Fusion

Finger pinching and imagination classification: A fusion of CNN architectures for IoMT-enabled BCI applications

A Brain–Computer Interface (BCI), integrated with the Internet of Medical Things (IoMT) and based on electroencephalogram (EEG) technology, allows users to control external devices by decoding brainwave patterns. Advanced deep learning-based BCIs, especially those utilizing sensorimotor rhythms (SMRs), have emerged as direct brain–device communication facilitators. SMRs involve users imagining limb motions to induce specific brain activity changes in the motor cortex. Despite progress, some users struggle with BCIs due to weak signals, individual variability, and limited task applicability. This study introduces an unsupervised EEG preprocessing pipeline for SMR-based BCIs. It evaluates an EEG dataset recorded during finger movements, employing two cleaning methods: an investigator-dependent pipeline and our proposed unsupervised method. Two distinct feature datasets are generated: one from cleaned EEG data processed into spectrogram images using supervised preprocessing, and another from data cleaned using our proposed unsupervised pipeline. The study extensively assesses five transfer learning convolutional neural network (TL-CNN) models for distinguishing Motor Imagery (MI) from finger movements (Mex) using the generated datasets. A novel probability fusion technique is developed to enhance TL-CNN classification in Mex versus MI finger-pinching actions. Comparative results show that the fusion-based method outperforms other state-of-the-art methods when applied to unsupervised EEG data. Specifically, our proposed approach achieves 97.9% accuracy, 93.4% precision, 95% recall, and an F1-score of 93.2%, demonstrating significant progress in distinguishing MI and Mex activities through the use of our unsupervised pre-processing pipeline and fusion-based CNN method. Our findings demonstrate the potential of our approach to serve as a benchmark for the global interdisciplinary research community and enable the development of future more effective and user-friendly real-time BCI systems.

Probabilistic Fusion Model for Industrial Soft Sensing Based on Quality-Relevant Feature Clustering

For most modern industrial processes with strong nonlinear and multimodal characteristics, the traditional linear PLS-based soft sensor may not work well. Meanwhile, the traditional global modeling approach has a high demand for data representation capability in the face of complex data distribution, which poses a challenge to soft sensing. In addition, the unbalanced nature of data distribution exacerbates the model's neglect of local information to some extent, which enhances the overall prediction difficulty of the model. To this end, based on the PLS, a novel quality-relevant feature clustering (QRFC) model is proposed for the first time in this paper from the view of local modeling of probabilistic fusion. In the QRFC, the PLS can give reasonable and explanatory guidance on the initial feature space for the modeling. Besides, the different data distributions are modeled using a unified perspective through a balanced grouping of the data. Further, a regulation variable is introduced to learn the feature space and complete the output-relevant clustering by iterative approach, to implicitly construct the correlation with the quality variable, forming a local approximation of the overall prediction capability. In general, the benefits of this framework, including better consistency of correlation and stronger abilities to deal with process nonlinearities and multimodality, lead to superior performance. To evaluate the feasibility and efficiency of the developed soft sensor, a real industrial case is taken as a demonstration. The experimental results demonstrate that the proposed method outperforms several other soft sensing approaches.

Temporally consistent reconstruction of 3D clothed human surface with warp field

Implicit functions are widely used in 3D human surface reconstruction due to their advantage to represent details. However, human reconstruction based on implicit functions struggles to maintain the integrity (unbroken body structure) and accuracy (no non-human parts) of human models. To address these issues, we propose a method, called TCR, for temporally consistent reconstruction of 3D clothed human surface with warp field. The fact that the general shape of a person does not change largely over time inspires us to exploit the temporally consistent shape information from previous frames to refine the human model of current frame. Therefore, we construct a canonical space and then store the shape information by updating the canonical model. To align the observed space with the canonical space, a warp field is firstly estimated for the forward and inverse warping of the human model. A probabilistic fusion strategy is then used to update the canonical model. In addition, the reconstructed result is further refined through the orthogonality constraints between the surface and its normal, which fully exploits the detailed information of estimated normal maps. Experiments on the Adobe and MonoPerfCap datasets show that TCR achieves the state-of-the-art performance. Furthermore, TCR is more robust and can maintain the integrity and accuracy of the reconstructed human body even with extreme poses and partial occlusions.

Optimization enabled deep learning approach with probabilistic fusion for image inpainting

ABSTRACT Nowadays, various deep learning (DL) approaches have been devised for image inpainting, which provided a substantial improvement in image quality. However, these approaches have failed to reconstruct the accurate structure of the original image. Hence, this research devised a novel and effective image inpainting approach, namely Autoregressive Flower Pollination Student Psychology Optimization (ArFPSPO). The image inpainting is carried out based on a newly devised hybrid context DL with the hybrid optimization scheme. The designed hybrid context DL approach adopts three DL techniques, such as Context encoder (CE), Context-Conditional Generative Adversarial Networks (CC-GAN), and Partial convolutional layer to complete the image inpainting process so that the outcomes are fused using probabilistic fusion with maximum entropy, thereby the final inpainted image is attained. Each of these three DL techniques is separately trained using a developed hybrid optimization technique. The experimental outcome reveals that the devised model gives optimal performance.

Reconstruction of 3D Reservoir Lithological Model Using 2D Facies Profiles in SU 36-11 Area of Ordos Basin, China

In the middle and late stages of gas field development, the establishment of a fine reservoir lithological model is an important basis for drilling well pattern adjustment and potential exploitation. The SU 36-11 area of the Ordos basin in China is developing braided channel sediment with rich gas resources. However, the success rate of drilling wells is low due to the complex reservoir heterogeneity and the lack of a fine reservoir lithological model. In this paper, the complex internal structure of the reservoir sand body is revealed using the architectural element analysis method. Three sand body models, that is, isolated channel, superimposed channel, and cut superimposed channel, can be recognized. The effective sand body is mainly the channel bar deposit with a thickness of 2–5 m, a width of 200–500 m, a length of 400–700 m, a width ratio of 50–120, and a length-to-width ratio of 1.5–2. The 2D maps of the lithofacies (architectural elements) were then digitized to create 2D training images (TI) for the construction of the 3D model. The 2D data template was selected to scan the TI to obtain the 2D multi-point probability. The 3D multi-point probability was then generated using the probability fusion theory. The Monte Carlo sampling was used to predict the lithological type between wells. Finally, the 3D reservoir lithological model was built directly using the 2D lithological profiles. From the model, the geometry of the braided channel, channel bar, and flood plain was well revealed, and the spatial distribution of effective reservoir sand bodies was accurately predicted. The cross-validation test shows that the error of the channel bar is 6.5% on average, which improves the accuracy of the prediction of lithology in the sub-surface and can be used to guide the subsequent development of residual gas.

Research on obstacle avoidance motion planning method of manipulator in complex multi scene

In order to improve the efficiency and success rate of obstacle avoidance motion planning of industrial manipulator in complex multi scenes, a collision detection model between manipulator and obstacles based on cylinder and sphere bounding box is established, and an improved RRT* algorithm based on heuristic probability fusion artificial potential field method(P-artificial potential field RRT*, PAPF-RRT*) is proposed. The probability target bias and random sampling point optimization strategy are introduced into the sampling, and the location optimization constraints are applied to the sampling points to enhance the sampling guidance and quality. In order to change the expansion direction of the traditional new node and the local optimization problem in special environment, the target gravity, obstacle repulsion and adaptive step size of the artificial potential field method are combined, so that the algorithm can guide the expansion direction and step size of the new node in real time within the resultant force range generated by APF, reducing excessive exploration and the expansion of the collision region. The Cubic B-spline is used to interpolate and optimize the planned path to reduce the complexity of the path and improve the flexibility of the path. The simulation results in two-dimensional and three-dimensional multi scenes show that the present algorithm reduces the average path search time by 31.22% and shortens the path length by 17.32% comparing with the traditional RRT* algorithm. The visual simulation results show that the present algorithm can make the manipulator successfully avoid obstacles and run to the target point quickly and smoothly.

IPos-5G: Indoor Positioning via Commercial 5G NR CSI

The fifth-generation (5G) networks have been massively deployed in commerce. The new features introduced by 5G networks are beneficial to wireless positioning. In this study, the performance of indoor positioning with commercial 5G new radio (NR) signals is investigated, and the channel state information (CSI) extracted from the downlink synchronization signal block is utilized. Considering the limited 5G NR base station (known as gNodeB) is hearable indoors, the fingerprint method is used, and an indoor positioning system termed iPos-5G is developed. The system consists of four components. Firstly, a module of quality control is applied for CSI preprocessing. Secondly, an unsupervised deep-autoencoder network is utilized to reconstruct CSI features. Thirdly, by supervised learning, a radial basis function is improved to optimize the probability model for similarity calculations. Finally, an amplitude-phase probability fusion function is proposed for positioning by weighting the coordinates of reference points. To verify the effectiveness of iPos-5G, indoor field tests are carried out in the scenarios of an office and a corridor. The test results show that iPos-5G achieves mean absolute errors of 2.14 and 2.81 m and standard deviation of the errors of 1.07 and 1.66 m, which outperforms the compared CSI fingerprint methods in terms of positioning accuracy and stability.

Adaptive rate image compressive sensing based on the hybrid sparsity estimation model

In some actual image compressive sensing (CS) systems, the sparsity of the original image signals is unknown on the sampling side. Resources on the sampling side are also limited. Therefore, it is a challenging task to implement adaptive rate allocation on the sampling side of a CS system. In this paper, a novel adaptive rate allocation block CS scheme for images based on a hybrid sparsity estimation model is proposed. First, a cosine similarity sparsity estimation model is constructed by directly modeling the measurement results, and on the reconstruction side, a thresholding cosine transform sparsity estimation model is constructed according to the initial reconstructed image. Then, the sparsity estimates generated by the above two models are fused according to their respective characteristics using a Bayesian probability fusion algorithm. The advantage of the proposed scheme is that it can significantly improve the accuracy of the final sparsity estimates while limiting the computational complexity on the sampling side, which makes it exceptionally suitable for wireless sensor network applications. On the reconstruction side, we propose an energy-based adaptive global reconstruction model, that can further reduce block artifacts caused by block-based CS. The experiment results show that, compared with previous state-of-the-art schemes, the proposed scheme can achieve a significant improvement in the peak signal-to-noise ratio at the same sampling rate.