Statistical Similarity Measures Research Articles

Wasserstein Distance-Assisted Variational Bayesian Kalman Filter With Statistical Similarity Measure for GNSS/INS Integrated Navigation

Wasserstein Distance-Assisted Variational Bayesian Kalman Filter With Statistical Similarity Measure for GNSS/INS Integrated Navigation

How Does ChatGPT Use Source Information Compared With Google? A Text Network Analysis of Online Health Information.

How Does ChatGPT Use Source Information Compared With Google? A Text Network Analysis of Online Health Information.

An invariant error-based SINS/DVL/USBL tightly coupled integrated navigation algorithm and its robust state estimator

The existing strapdown inertial navigation system (SINS)/Doppler velocity log (DVL)/ultra-short baseline system (USBL) integrated navigation error models directly apply algebraic differences to define the vector error, which is not strictly accurate. Considering that the invariant error definitions derived from Lie group theory have been used in the navigation field to obtain more accurate linear error models, in this paper, a new SINS/DVL/USBL integrated navigation system is investigated based on the invariant error definition and tightly coupled structure. In order to suppress the effect of measurement outliers on navigation accuracy, a new statistical similarity measure (SSM)-based multiple adaptive outlier-robust state estimator (MAORSE) is proposed. It decomposes the noise covariance matrix into multiple sub-matrices. Through the estimation and adjustment of sub-matrices, the influence of abnormal observations on the filtering results is weakened while realizing the full utilization of normal observation information. The test results demonstrate the effectiveness and superiority of the proposed invariant error-based SINS/DVL/USBL tightly coupled integrated navigation algorithm and SSM-based MAORSE.

A framework for evaluating the performance of SMLM cluster analysis algorithms.

Single-molecule localization microscopy (SMLM) generates data in the form of coordinates of localized fluorophores. Cluster analysis is an attractive route for extracting biologically meaningful information from such data and has been widely applied. Despite a range of cluster analysis algorithms, there exists no consensus framework for the evaluation of their performance. Here, we use a systematic approach based on two metrics to score the success of clustering algorithms in simulated conditions mimicking experimental data. We demonstrate the framework using seven diverse analysis algorithms: DBSCAN, ToMATo, KDE, FOCAL, CAML, ClusterViSu and SR-Tesseler. Given that the best performer depended on the underlying distribution of localizations, we demonstrate an analysis pipeline based on statistical similarity measures that enables the selection of the most appropriate algorithm, and the optimized analysis parameters for real SMLM data. We propose that these standard simulated conditions, metrics and analysis pipeline become the basis for future analysis algorithm development and evaluation.

Fusing ToF‐SIMS Images for Spatial‐Spectral Resolution Enhancement using a Convolutional Neural Network

AbstractHyperspectral data sets generated by time‐of‐flight secondary ion mass spectrometry (ToF‐SIMS) contain valuable spatial‐spectral information characterizing the distribution of atomic and molecular species across a sample surface. Modern ToF‐SIMS instruments have high spatial resolution (in the order of tens of nanometers) relative to most other mass spectrometry imaging (MSI) techniques. However, there is generally a trade‐off between spatial and mass resolution when using different instrument modes. In this study, a convolutional neural network (CNN) fusion method is used to fuse correlated high spatial and high mass resolution ToF‐SIMS hyperspectral data sets. This process generates resolution‐enhanced data, which exhibit both high spatial and mass resolution. The CNN fusion method is applied to ToF‐SIMS images of a simple, well‐characterized gold mesh sample and a significantly more complex biological (tumor) tissue section. The method is compared to another linear fusion method used in the broader MSI community and a substantial improvement is found. This comparison focuses on both visual quality observations as well as statistical similarity measures. This work demonstrates the utility of the CNN fusion method for ToF‐SIMS data, enabling investigation of the atomic and molecular characteristics of surfaces at high spatial and mass spectral resolution.

A Novel DVL Calibration Method Based on Robust Invariant Extended Kalman Filter

The calibration accuracy of the Doppler Velocity Log (DVL) error directly affects the performance of the SINS/DVL integrated navigation system. Previous studies of DVL calibration have not dealt with DVL measurement outliers. Therefore, in this paper, a novel DVL calibration method is proposed to reduce the influence of outliers on calibration accuracy. Introducing the Lie group theory to the DVL calibration study for the first time, and taking advantage of the Special Orthogonal Group of order 3 [SO(3)] representation, a new DVL calibration model based SO(3) group with the ability to represent large misalignment angles is established. And an invariant extended Kalman filter (IEKF) design for the DVL calibration model is introduced. Then, a robust IEKF is proposed by combining the linear error propagation equation of the SO(3)-based DVL calibration model with the statistical similarity measure (SSM) theory. The proposed robust IEKF algorithm enriches the IEKF theory and has the ability to accurately estimate the state of the Lie group model in outlier environments. The simulation and lake trial are used to illustrate the effectiveness and superiority of the proposed DVL calibration method.

Statistical Similarity Measure-Based Adaptive Outlier-Robust State Estimator With Applications

This article presents an adaptive outlier-robust state estimator (AORSE) under the statistical similarity measures (SSMs) framework. Two SSMs are first proposed to evaluate the similarities between a pair of positive definite random matrices and between a pair of weighted random vectors, respectively. The AORSE is developed by maximizing a hybrid SSMs based cost function, wherein the posterior density function of the hidden state is assumed as a Gaussian distribution with the posterior covariance being approximately determined in a heuristic way. Simulation and experimental examples of moving-target tracking demonstrate the effectiveness of the proposed algorithm.

Autonomous Prostate Segmentation in 2D B-Mode Ultrasound Images

Prostate brachytherapy is a treatment for prostate cancer; during the planning of the procedure, ultrasound images of the prostate are taken. The prostate must be segmented out in each of the ultrasound images, and to assist with the procedure, an autonomous prostate segmentation algorithm is proposed. The prostate contouring system presented here is based on a novel superpixel algorithm, whereby pixels in the ultrasound image are grouped into superpixel regions that are optimized based on statistical similarity measures, so that the various structures within the ultrasound image can be differentiated. An active shape prostate contour model is developed and then used to delineate the prostate within the image based on the superpixel regions. Before segmentation, this contour model was fit to a series of point-based clinician-segmented prostate contours exported from conventional prostate brachytherapy planning software to develop a statistical model of the shape of the prostate. The algorithm was evaluated on nine sets of in vivo prostate ultrasound images and compared with manually segmented contours from a clinician, where the algorithm had an average volume difference of 4.49 mL or 10.89%.

Geodiversity assessment with global and local spatial multicriteria analysis

The paper reports on geodiversity assessments of three national parks in Poland, performed independently by experts and volunteers. Each national park selected for the study represents one of three morphogenetically different landscape types: mountains, uplands, and plains. The expert and volunteer assessment data sets were separately processed with two spatial multicriteria methods: Weighted Linear Combination (WLC) - also referred to in the paper as the global version of WLC, and Local Weighted Linear Combination (L-WLC) resulting in two geodiversity maps for each of the parks – one based on WLC and the other on L-WLC output data. The maps were qualitatively evaluated for their efficacy of capturing spatial heterogeneity and differentiating between high and low geodiversity of specific areas within the national parks. The expert-based maps were compared with the volunteer-based maps using statistical measures of association and similarity. The results show that L-WLC is more suitable for geodiversity mapping of mountainous areas characterized by high morphogenetic and morphometric diversity whereas WLC yields better results in less diverse areas such as uplands and lowlands. The use of data originating from volunteer-based assessment requires meeting internal and external data quality standards and should be treated with caution.

Sparsity Regularized Deep Subspace Clustering for Multicriterion-Based Hyperspectral Band Selection

Hyperspectral images provide rich spectral information corresponding to visible and Near-infrared imaging (NIR) regions, facilitating accurate classication, object identication, and target detection. However, the high volume of data creates a computational challenge in processing. The band selection process identies specic informative and discriminative spectral bands from the data speed up the processing without impeding the performance. This paper presents an application-independent band selection framework that utilizes improved sparse deep subspace clustering and introduces an efcient multicriteria-based representative band selection (BS). The proposed sparse deep subspace clustering approach efciently identies the underlying non-linear subspace structure of the data and organizes the data accordingly. The work introduces a novel, robust sparsity measure to obtain a powerful self-representation and ameliorated performance compared to the prevalent subspace clustering methods. The work subsequently selects the representative bands from each cluster by combining structural information of the band images with the statistical similarity measure. We evaluate the BS performance on standard real images using information-theoretic criterion, classication, and unmixing performance. The comparative performance assessment demonstrates that our proposed method identies the informative bands and out.

A Robust SINS/USBL Integrated Navigation Algorithm Based on Earth Frame and Right Group Error Definition

In this paper, a new SINS/USBL integrated navigation algorithm is proposed based on the Earth-Centered Earth-Fixed frame SINS mechanization. The dynamic model of the navigation system has proven to satisfy the “group affine” property. Using the Lie group theory to define the right group state error vector, a new SINS / USBL error model is developed to accurately describe the error propagation under error-polluted navigation parameters. The most significant advantage of this error model is its ability to handle large initial misalignment angles. Subsequently, considering the low update frequency of the USBL measurement update, a virtual observation equation is constructed to improve the convergence speed of the algorithm in presence of large initial misalignment angles. Finally, for the sake of effectively suppressing the influence of acoustic outliers, the existing outlier Kalman filter based on statistical similarity measure is improved by adaptively adjusting the dof parameters, so that inaccurate dof parameters have less impact on filter performance. The effectiveness and superiority of the proposed algorithm are verified by simulation and sea trial.

An Improved Robust Kalman Filter for SINS/DVL Tightly Integrated Navigation System

Aiming at the problem of outliers in Doppler velocity log (DVL) beam measurements, for strapdown inertial navigation system (SINS)/Doppler velocity log (DVL) tightly coupled system, an improved robust Kalman filter (KF) algorithm is proposed, which can handle the measurement noise variance of each DVL beam individually. First, the multidimensional measurement equation is decomposed into several 1-D measurement equations, i.e., different DVL beam measurements are assumed to be obtained by different sensors. Second, the statistical similarity measure (SSM) is introduced to quantify the similarity between two random vectors. And a new cost function based on SSM is constructed for the dynamic system containing one state equation and multiple measurement equations. To optimize this cost function, the iterative procedures to update the posterior state is derived based on the Gauss-Newton iteration algorithm. Both the simulation and lake trial illustrate the superiority and effectiveness of the proposed algorithm. It is shown that in the case of partial DVL beams with outliers, the proposed algorithm can avoid the loss of normal measurement information and thus improve estimation accuracy.

A Novel Adaptive Filter for Cooperative Localization Under Time-Varying Delay and Non-Gaussian Noise

In this study, to cope with time-varying delay in acoustic communication under non-Gaussian noise caused by outliers, a novel robust delay filtering algorithm is proposed for cooperative localization of autonomous underwater vehicles (AUVs). First, the modified measurement equations of the nonlinear cooperative location system with time-varying delay are derived, and the delay caused by the information processing and propagation of the underwater acoustic modem is converted into measurement bias. Second, to improve the robustness of the system with outliers caused by abnormal measurement values of underwater acoustic communication and Doppler velocity log (DVL), the statistical similarity measure (SSM) is introduced to construct the cost function. This is accomplished by quantifying the similarity between the state vector and the predicted state vector, similarity between the delay measurement, and modified predicted delay measurement. Finally, the modified measurement noise variance is obtained by maximizing the cost function, and two algorithms for the approximate solution of the nonlinear cost function are presented. The proposed robust delayed algorithm alleviates the impacts of delayed measurement with outliers on localization accuracy. The effectiveness and potential of the proposed filter are verified by lake trials.

A novel statistical mechanics-based metric for characterization of text documents

Machine learning (ML) uses intelligence-based statistical models to resolve characterization problems. However, the accuracy of ML models decreases as the volume of data increases since they are not able to capture the diversity and important regularities accompanying the huge volume. In this regard, the statistical mechanics’ (SM) paradigm for ML models must be investigated. SM like Big Data also deals with a large volume of instances. A correlation principle is proposed between data represented as words in a text corpus and a large volume of particles within a system. Based on it a new similarity measure, namely degeneracy-based statistical similarity measure (DSSM) is proposed. The DSSM is directly proportional to the ratio of the total energy of two documents with maximally similar words and the total number of distinct words in the document. The DSSM showed higher performance with regards to existing measures such as Euclidean, Cosine, SMTP, and MBSM.

A Novel Outlier-Robust Kalman Filtering Framework Based on Statistical Similarity Measure

In this article, a statistical similarity measure is introduced to quantify the similarity between two random vectors. The measure is, then, employed to develop a novel outlier-robust Kalman filtering framework. The approximation errors and the stability of the proposed filter are analyzed and discussed. To implement the filter, a fixed-point iterative algorithm and a separate iterative algorithm are given, and their local convergent conditions are also provided, and their comparisons have been made. In addition, selection of the similarity function is considered, and four exemplary similarity functions are established, from which the relations between our new method and existing outlier-robust Kalman filters are revealed. Simulation examples are used to illustrate the effectiveness and potential of the new filtering scheme.

Visualizing public transit system operation with GTFS data: A case study of Calgary, Canada

Public transportation agencies are one of the industries that generate a large volume of data on a high frequency and velocity basis. The General Transit Feed Specification (GTFS) is one of the datasets these agencies generate and share openly with the public. GTFS feeds contain data for scheduled transit service including stop and route locations, and schedules information. This paper aims to demonstrate the potential of GTFS data, specifically, the paper describes the development of a GTFS data visualization tool that displays spatial and temporal patterns of transit services from which qualitative information and insights can be gained. In this paper, GTFS data from Calgary Transit was used as a case study. Previous studies focused on the development of visualization tools that display transit movement, or static graphical representation of transit operation. However, there is still a need for a dynamic interactive visualization tool that can also measures and displays transit system operation geographically and statistically. This study builds on the previous investigations and further develops a new public transit system operation visualization tool (called PubtraVis) with six visualization modules that reflect on different transit system operational characteristics; mobility, speed, flow, density, headway, and analysis. The user can evaluate two modules side by side for comparative analysis. The analysis module provides an insightful statistical summary and similarity measure and clustering results based on the transit operation characteristics. PubtraVis was tested with real-world users through a user experience study from which it was found to be useful and easy to start using. PubtraVis can be a useful tool to demonstrate the dynamism of transit vehicles from the entire transit network at a glance, and can be used to facilitate communication between transit operators, city authorities, and the general public regarding the public transit planning and operation.

Region-Merging Method With Texture Pattern Attention for SAR Image Segmentation

This letter proposes a region-merging-based method for synthetic aperture radar (SAR) image segmentation, where the merging cost is a fusion of texture pattern similarity measure (TPSM), statistical similarity measure (SSM), and the relative common boundary length penalty (RCBLP). The segmentation is implemented in three steps. First, an image is oversegmented based on the multiscale Bhattacharyya distance to generate an initial partition of considerable regions. Second, regions with sizes under a given threshold are mandatorily merged to yield a middle segmentation. Third, a region-merging process using the new merging cost is iteratively conducted to achieve final segmentation. Due to the existence of the TPSM in the merging cost, the new method avoids the false merging of adjacent regions with different textures. Experimental results of the real SAR images show that the proposed method outperforms existing region-merging-based methods.

Assessment of CO2‐enhanced oil recovery and associated geologic storage potential in the Michigan Northern Pinnacle Reef Trend

AbstractThis paper provides an improved estimate of CO2‐EOR (where CO2 is carbon dioxide and EOR is enhanced oil recovery) and CO2 storage potential for depleted oil fields in Michigan's northern pinnacle reef trend (NPRT). Our methodology is based on capturing data on reservoir performance from reefs currently undergoing CO2‐EOR operations in the NPRT (referred to as ‘monitored reefs’), and then applying them to other reefs within the NPRT (referred to as ‘catalog reefs’). For each monitored reef, we calculate fractional primary recovery, fractional incremental EOR recovery, net utilization ratio, and storage efficiency factor. The corresponding incremental oil recovery from EOR, storage capacity until end of EOR, and total CO2 injection needs are then estimated for each catalog reef and combined (over all monitored reefs) using a weighted averaging procedure. These weights are related to a statistical similarity measure that is calculated between each monitored reef and each catalog reef based on a number of variables related to production data, formation type, or descriptive geologic attributes. For the entire NPRT catalog of 383 reefs as used in this study, our results indicate 118 million (MM) stock tank barrels (STB) (1.88 × 107 Sm3) of incremental oil from EOR operations, corresponding to 49 MM metric tons (MT) of CO2 storage and 266 MM MT of total CO2 injection. However, approximately one‐third of the reefs provide two‐thirds of the potential for CO2‐EOR and geologic storage, assuming an economic threshold of 0.5 MM STB (80 000 Sm3) of incremental oil from EOR. © 2020 The Authors. Greenhouse Gases: Science and Technology published by Society of Chemical Industry and John Wiley & Sons, Ltd.

Detecting Paraphrases in Marathi Language

Paraphrasing refers to the sentences that either differs in their textual content or dissimilar in rearrangement of words but convey the same meaning. Identifying a paraphrase is exceptionally important in various real life applications such as Information Retrieval, Plagiarism Detection, Text Summarization and Question Answering. A large amount of work in Paraphrase Detection has been done in English and many Indian Languages. However, there is no existing system to identify paraphrases in Marathi. This is the first such endeavor in Marathi Language. A paraphrase has different structured sentences and Marathi being semantically strong language hence this system is designed for checking both statistical and semantic similarity of Marathi sentences. Statistical similarity measure does not need any prior knowledge as it is only based on the factual data of sentences. The factual data is calculated on the basis of the degree of closeness between the word-set, word-order, word-vector and word-distance. Universal Networking Language (UNL) speaks about the semantic significance in the sentence without any syntacticpointofinterest.Hence, these mantic similarity calculated on the basis of generated UNL graphs for two Marathi sentences renders semantic equality of two Marathi sentences. The total para phrases core was calculated after joining statistical and semantic similarity scores which gives the judgement of being paraphrase or non-paraphrase about the Marathi sentences.

Detecting paraphrases in the Marathi language

Paraphrasing refers to writing that either differs in its textual content or is dissimilar in rearrangement of wordsbut conveys the same meaning. Identifying a paraphrase is exceptionally important in various real life applicationssuch as Information Retrieval, Plagiarism Detection, Text Summarization and Question Answering. A large amountof work in Paraphrase Detection has been done in English and many Indian Languages. However, there is noexisting system to identify paraphrases in Marathi. This is the first such endeavor in the Marathi Language.A paraphrase has differently structured sentences, and since Marathi is a semantically strong language, thissystem is designed for checking both statistical and semantic similarities of Marathi sentences. Statistical similaritymeasure does not need any prior knowledge as it is only based on the factual data of sentences. The factual datais calculated on the basis of the degree of closeness between the word-set, word-order, word-vector and word-distance. Universal Networking Language (UNL) speaks about the semantic significance in sentence without anysyntactic points of interest. Hence, the semantic similarity calculated on the basis of generated UNL graphs for twoMarathi sentences renders semantic equality of two Marathi sentences. The total paraphrase score was calculatedafter joining statistical and semantic similarity scores, which gives a judgment on whether there is paraphrase ornon-paraphrase about the Marathi sentences in question.