Mean Shift Vector Research Articles

A mean shift algorithm incorporating reachable distance for spatial clustering

Spatial clustering is a widely used technique in spatial analysis that groups similar objects together based on their proximity in space. However, traditional clustering algorithms often fail to ensure the accessibility of cluster centers, which limits their validity in practical applications such as facility location problems. To address this issue, this article introduces a novel Mean Shift algorithm that incorporates reachable distance and an iterative mechanism to accurately locate cluster centers. The proposed algorithm initially labels clustering elements with road network coordinates to facilitate the calculation of reachable distance and the cluster center iterative mechanism. Subsequently, the mean shift vector function is modified to employ reachable distance as the measure of geographic reachable similarity. Unlike existing algorithms, our approach allows for cluster centers to be positioned independently of the clustering elements, guaranteeing geographical accessibility. Through simulation experiments, we demonstrate that our proposed algorithm not only outperforms existing methods in terms of solution quality, but also effectively addresses the limitations of disregarding geographical obstacles and unreachable cluster centers.

EMS: A Large-Scale Eye Movement Dataset, Benchmark, and New Model for Schizophrenia Recognition.

Schizophrenia (SZ) is a common and disabling mental illness, and most patients encounter cognitive deficits. The eye-tracking technology has been increasingly used to characterize cognitive deficits for its reasonable time and economic costs. However, there is no large-scale and publicly available eye movement dataset and benchmark for SZ recognition. To address these issues, we release a large-scale Eye Movement dataset for SZ recognition (EMS), which consists of eye movement data from 104 schizophrenics and 104 healthy controls (HCs) based on the free-viewing paradigm with 100 stimuli. We also conduct the first comprehensive benchmark, which has been absent for a long time in this field, to compare the related 13 psychosis recognition methods using six metrics. Besides, we propose a novel mean-shift-based network (MSNet) for eye movement-based SZ recognition, which elaborately combines the mean shift algorithm with convolution to extract the cluster center as the subject feature. In MSNet, first, a stimulus feature branch (SFB) is adopted to enhance each stimulus feature with similar information from all stimulus features, and then, the cluster center branch (CCB) is utilized to generate the cluster center as subject feature and update it by the mean shift vector. The performance of our MSNet is superior to prior contenders, thus, it can act as a powerful baseline to advance subsequent study. To pave the road in this research field, the EMS dataset, the benchmark results, and the code of MSNet are publicly available at https://github.com/YingjieSong1/EMS.

Outlier detection in spatial error models using modified thresholding-based iterative procedure for outlier detection approach

BackgroundOutliers, data points that significantly deviate from the norm, can have a substantial impact on statistical inference and provide valuable insights in data analysis. Multiple methods have been developed for outlier detection, however, almost all available approaches fail to consider the spatial dependence and heterogeneity in spatial data. Spatial data has diverse formats and semantics, requiring specialized outlier detection methodology to handle these unique properties. For now, there is limited research exists on robust spatial outlier detection methods designed specifically under the spatial error model (SEM) structure.MethodWe propose the Spatial-Θ-Iterative Procedure for Outlier Detection (Spatial-Θ-IPOD), which utilizes a mean-shift vector to identify outliers within the SEM. Our method enables an effective detection of spatial outliers while also providing robust coefficient estimates. To assess the performance of our approach, we conducted extensive simulations and applied it to a real-world empirical study using life expectancy data from multiple countries.ResultsSimulation results showed that the masking and JD (Joint Detection) indicators of our Spatial-Θ-IPOD method outperformed several commonly used methods, even in high-dimensional scenarios, demonstrating stable performance. Conversely, the Θ-IPOD method proved to be ineffective in detecting outliers when spatial correlation was present. Moreover, our model successfully provided reliable coefficient estimation alongside outlier detection. The proposed method consistently outperformed other models (both robust and non-robust) in most cases. In the empirical study, our proposed model successfully detected outliers and provided valuable insights in the modeling process.ConclusionsOur proposed Spatial-Θ-IPOD offers an effective solution for detecting spatial outliers for SEM while providing robust coefficient estimates. Notably, our approach showcases its relative superiority even in the presence of high leverage points. By successfully identifying outliers, our method enhances the overall understanding of the data and provides valuable insights for further analysis.

Flexible multi-objective particle swarm optimization clustering with game theory to address human activity discovery fully unsupervised

Human activity recognition is a crucial field of study, but current approaches often require ground truth labels, which are not always available. We propose a new method called the Flexible Multi-Objective Particle swarm optimization clustering method based on Game theory (FMOPG), which can identify human activities without any supervision. Unlike traditional clustering methods that require an estimate of the number of clusters and are often inaccurate, FMOPG handles varying cluster numbers with an incremental technique, selecting clusters with good connectivity and separation. We enhance Particle Swarm Optimization (PSO) with mean-shift vectors for faster convergence and better handling of non-spherical clusters. Employing multi-objective optimization and Gaussian mutation, FMOPG provides a range of optimal solutions. We map the optimization problem to game theory to select the best solution based on different criteria. A smart grid-based method is proposed for population initialization, reducing variance and improving reliability. FMOPG outperforms state-of-the-art methods, improving clustering accuracy by 3.65%. Moreover, the incremental technique has improved clustering time by 71.18%.

Web Mapping for Farm Management Information Systems: A Review and Australian Orchard Case Study

A web mapping XYZ Tile Layer Service, such as Google Earth (GE), provides an amazing resource for the visualization of spatial data against aerial and satellite imagery with global coverage, typically at a resolution finer than 5 m. However, the increasing requirement on spatial accuracy in farm information requires a greater appreciation of the issues involved in the use of such services. Position errors can be created in the georeferencing and orthorectification of images, transformation between reference frames (datums) in map projection, e.g., using a spheroid as compared to an ellipsoid earth model, and tectonic shifts. A review is provided of these issues, and a case study is provided of the horizontal positional accuracy of web map imagery for Australian mango orchards. Positional accuracies varied from 1.804 to 6.131 m across four farms using GE 2021 imagery, between 1.556 and 3.365 m in one farm for the most recent imagery available from each of four web map providers, and from 0.806 m (in 2016) to 10.634 m (in 2003) in one farm for the period of 2003 and 2021 using the historical GE imagery resource. A procedure involving the estimation of four transformation parameters was demonstrated for the alignment of GNSS data with GE imagery. However, as the scale factor was unity and the rotational value was near zero, the use of a simple horizontal mean shift vector was recommended. Further recommendations are provided on (i) the use of web mapping services, with a comparison of the use of UAV survey imagery, and (ii) the need for metadata, particularly the date of collection, on collected position data, in the context of use in farm management information systems.

Antiocclusion Visual Tracking Algorithm Combining Fully Convolutional Siamese Network and Correlation Filtering.

Machine learning only uses single-channel grayscale features to model the target, and the filter solution process is relatively simple. When the target has a large change relative to the initial frame, the tracking effect is poor. When there is the same kind of target interference in the target search area, the tracking results will be poor. The tracking algorithm based on the fully convolutional Siamese network can solve these problems. By learning the similarity measurement function, the similarity between the template and the target search area is evaluated, and the target area is found according to the similarity. It adopts offline pre-training and does not update online for tracking, which has a faster tracking speed. According to this study, (1) considering the accuracy and speed, the target tracking algorithm based on correlation filtering performs well. A sample adaptive update model is introduced to eliminate unreliable samples, which effectively enhances the reliability of training samples. With simultaneous changes in illumination and scale, fast motion and in-plane rotation IPR can still be maintained. (2) Determined by calculating the Hessian matrix, in the Struck function, Bike3 parameter adjustment can achieve fast tracking, and Boat5 ensures that the system stability is maintained in the presence of interference factors. The position of the highest scoring point in the fine similarity score map of the same size as the search image is obtained by bicubic interpolation as the target position. (3) The parallax discontinuity caused by the object boundary cannot be directly processed as a smooth continuous parallax. The MeanShift vector obtained by calculating the target template feature and the feature to be searched can increase the accuracy by 53.1%, reduce the robustness by 31.8%, and reduce the error by 28.6% in the SiamVGG algorithm.

Fusion of multi-source retinal fundus images via automatic registration for clinical diagnosis

Diabetic retinopathy, age-related macular degeneration and glaucoma, are the leading causes of visual impairment or blindness of the population across different ages. Retinal fundus imaging is a clinically regular tool for the diagnosis of retinal diseases. In the interest of having a comprehensive understanding of the fundus condition, it is valuable to leverage multiple fundus images from different modalities. However, a direct fusion of the multi-source fundus images eases to mis-align the physiological structure or spatial position due to possible eyeball rotations or head movements. The problem turns out to be more severe if the images were corrupted by ill conditions on eyes, such as micro-bleeding and plaques. To tackle this problem, we propose a multi-source registration model for retinal fundus images. Our proposed method considers multiple correspondences and dual structural constraints during the registration process. The method firstly selects adequate feature points by an adjustable threshold selection strategy. Then a feature-guided correspondence estimation model is established to build complementary features. Finally, their spatial transformation is built by using mean shift evolution. The evolution is guided by Tikhonov regularization on dual geometric structures. It overcomes the mess of mean shift vector field and mitigating the ill-posed displacement in field recovery. We have conducted our method on the collected 220 multi-source retinal fundus image pairs, which involve minor and larger displacement or severe retinopathy lesions, as well as additive different intensities of Gaussian noises. Extensive experiments demonstrate that the proposed method consistently outperforms seven feature-based methods.

Подбор вектора признаков для процедуры MeanShift

The article gives the description of the feature vector, which is suitable for the MeanShift procedure, uses all the color information of the RGB24 format and has a dimension exceeding only 1.5 times the dimension of the smallest 512-dimensional vector used for the Kernel Based Object Tracking procedure. For the described feature vector, a function of similarity of two elliptical areas of the frame is built. For the similarity function, formulas are found for the gradient vector - the mean shift vector, which indicates the direction of the growth of similarity in four-dimensional space of all elliptical regions covering the object in the frame. Knowing the greatest value of the similarity function of two elliptical regions, the length of the displacement vector in the four dimensional space of all elliptical regions was found. To this vector the previous point in space must be moved at the current moment, i.e. the values of the coordinates of the center and the dimensions of the ellipse, in order to obtain the best similarity of the current elliptical area from the previous one. Finally, so as to implement Kernel Based Object Tracking, an algorithm of successive iterations (Newton's method) has been developed, which allows finding the parameters of the ellipse that really has the best similarity. The experiments were carried out and their results were presented and discussed

Individual Tree Segmentation Based on Mean Shift and Crown Shape Model for Temperate Forest

Light detection and ranging (LiDAR) provides high-resolution geometric information for monitoring forests at individual tree crown (ITC) level. An important task for ITC delineation is segmentation, and previous studies showed that the adaptive 3-D mean shift (AMS3D) algorithm provides effective results. AMS3D for ITC segmentation has three components for the kernel profile: shape, weight, and size. In this letter, we present an AMS3D approach based on the adaptation of the kernel profile size through an ellipsoid crown shape model. The algorithm parameters are estimated based on allometry equations derived from 22 forest plots in two study sites. After computing the mean shift (MS) vector, we initialize the parameters of the ellipsoid crown shape model to derive the kernel profile size, and further tested two crown shape models for adapting the size of the superellipsoid (SE) kernel profile. These schemes are compared with two other MS algorithms with and without kernel profile size adaptation. We select the best algorithm output per plot based on the maximum F1-score. The ellipsoid crown shape model with a SE kernel profile of n = 1.5 presents the highest recall and the best Jaccard index, especially for conifers.

Memory access minimization for mean-shift tracking in mobile devices

Due to the development of artificial intelligence and computer vision technology, many autonomous drones have been studied. However, computer vision technology requires high performance CPU due to its high complexity, and battery consumption is so high that drones are constrained to fly for a long time. Therefore, low-power mobile devices require tracking algorithms that minimize battery consumption. In this paper, we propose a mean-shift based tracking algorithm that minimizes memory access to reduce battery consumption. To accomplish this, we minimize the number of memory accesses by using an algorithm that divides the direction of the mean-shift vector into eight, and calculates the sum of the density maps only for the new area without calculating the sum of the density maps for the already calculated area. It is possible to increase the calculation efficiency by lowering the memory access cost. Experimental results show that the proposed method is more efficient than the existing method.

Fast and accurate mean-shift vector based wavelength extraction for chromatic confocal microscopy

Fast and accurate peak wavelength extraction of chromatic confocal signals is of vital importance for efficient chromatic confocal measurement. Existing peak extraction algorithms do not effectively balance calculation efficiency and extraction accuracy. For example, centroid algorithms have high computational efficiency but low accuracy; model-based fitting algorithms have high extraction accuracy but low calculation efficiency. In this paper, a fast and accurate mean-shift vector based peak wavelength extraction algorithm is proposed. In this algorithm, the initial peak wavelength is preset with a centroid algorithm; then the corresponding mean-shift vector is calculated, after which, a new peak wavelength can be calculated by shifting the peak wavelength along the mean-shift vector. This process is then iteratively carried out until a preset ending condition is met. Our experimental results show that the mean-shift vector based peak extraction algorithm has an accuracy equal to that achieved with Gaussian fitting while the efficiency improves by over 70 times.

BorderShift: toward optimal MeanShift vector for cluster boundary detection in high-dimensional data

We present a cluster boundary detection scheme that exploits MeanShift and Parzen window in high-dimensional space. To reduce the noises interference in Parzen window density estimation process, the kNN window is introduced to replace the sliding window with fixed size firstly. Then, we take the density of sample as the weight of its drift vector to further improve the stability of MeanShift vector which can be utilized to separate boundary points from core points, noise points, isolated points according to the vector models in multi-density data sets. Under such circumstance, our proposed BorderShift algorithm doesn’t need multi-iteration to get the optimal detection result. Instead, the developed Shift value of each data point helps to obtain it in a liner way. Experimental results on both synthetic and real data sets demonstrate that the F-measure evaluation of BorderShift is higher than that of other algorithms.

Importance Sampling for Credit Portfolio Risk with Risk Factors Having t-Copula

This paper proposes an efficient simulation method for calculating credit portfolio risk when risk factors have a heavy-tailed distributions. In modeling heavy tails, its features of return on underlying asset are captured by multivariate [Formula: see text]-Copula. Moreover, we develop a three-step importance sampling (IS) procedure in the [Formula: see text]-copula credit portfolio risk measure model for further variance reduction. Simultaneously, we apply the Levenberg–Marquardt algorithm associated with nonlinear optimization technique to solve the problem that estimates the mean-shift vector of the systematic risk factors after the probability measure change. Numerical results show that those methods developed in the [Formula: see text]-copula model can produce large variance reduction relative to the plain Monte Carlo method, to estimate more accurately tail probability of credit portfolio loss distribution.

Ordering for Reduced Transmission Energy Detection in Sensor Networks Testing a Shift in the Mean of a Gaussian Graphical Model

Detection of a shift in the mean of a vector following a decomposable Gaussian graphical model (DGGM) is considered, where each component of the vector is measured at a different sensor in a network. We provide a new method eliminating transmissions normally needed in an optimum clustering approach by using ordered transmissions while achieving the same Bayes risk as the optimum clustering approach. In the new approach, the Bayes optimum test statistic is represented as a sum of local test statistics, where each local test statistic depends only on the observations made at one clique in a generalization of the ordered transmission approach previously suggested for statistically independent observations. Hence, we propose to organize the sensors into clusters based on the clique of the DGGM they belong to, and each cluster selects one sensor to be its cluster head (CH). After collecting and summarizing the observed data at each cluster, the ordered transmission approach is employed over the CHs in an attempt to reduce the number of communications from the CHs to the fusion center where the decision is made. It is shown that the developed approach can guarantee a lower bound on the average number of transmissions saved from the ordered transmission approach for any given DGGM which approaches approximately half the number of cliques when the norm of the mean-shift vector in each clique becomes sufficiently large. In all the cases considered, numerical results imply that a significant portion of the transmissions can be saved, and the developed lower bound is obeyed. The development of the appropriate local processing and the proof of savings are highly nontrivial generalizations of ordering for statistically independent observations and this paper represents the first justification of an ordering approach for cases with statistically dependent observations.

Mean LBP and modified fuzzy C-means weighted hybrid feature for illumination invariant mean-shift tracking

Object tracking is a critical task in surveillance and activity analysis. Two main issues for tracking are appearance (illumination) and structural (size of a target) variations of the object. We propose a method which is robust and addresses these issues by incorporating features that are less variant to these changes. The proposed features are mean local binary pattern (mLBP), an illumination invariant texture feature, and modified fuzzy c-means (MFCM) weighted color histogram to handle both illumination and scale changes. These features are combined to form a hybrid mean-shift (MS) vector and used in the MS vector framework for target tracking. Experimental results using standard benchmark videos show that the proposed scheme can lead to better localization and robust tracking in challenging illumination scenarios, when compared to several existing tracking algorithms.

Improved FPGA Implementation of Real Time Modified Mean Shift Tracking Algorithm

Objectives: To modify and implement color based mean shift object detection and tracking algorithm utilizing both the parallel and sequential capabilities of Xilinx ZYNQ ZC-702 SoC in order to speed up tracking. Method/ Statistical Analysis: The parallel and sequential processing capabilities of Field Programmable Logic Array (FPGA) and Processing System (PS) respectively are utilized in order to have a standalone system that can be faster, reliable and efficient while tracking the object in real time. Operations such as reading and writing video, grabbing kernel from frame and mean shift vector computation are sequential in nature and are best suited for processing system where as the operations parallel in nature are best fit for FPGA and may include estimation of histogram, computation of weights and estimation of weighted histogram. Executing them in parallel helps in reducing the machine cycles and enhances the fps. Findings: The high computational power of the algorithm is met by collective use of hardware and software while keeping the resources available on FPGA in check. The modified mean shift tracking method helps in exploiting the parallel computation capability of the FPGA. The paper compares the results with various techniques implemented on different embedded boards and the frame processing rate is much better with proposed FPGA implementation of modified mean shift tracking algorithm. Further, the window size can be varied without affecting fps. Application/Improvements: The frame rate achieved by using both hardware and software simultaneously is considerably higher than achieved with earlier implementations.

BNIMS: Block-based Non-iterative Mean-shift Segmentation algorithm for Medical Images

This paper proposed a novel Block based Mean Shift Image Segmentation Algorithm to significantly reduce the computation and improve the segmentation accuracy for high resolution Medical Image. One of the challenging tasks in the image analysis and computer vision area is to correctly classify the pixels as there are no crisp borders among entities in an image. In this proposed methodology, it is observed that the computational complexity of the procedure is diminished by combining the pixels of an image of size MXN into non overlapping image blocks of size 3x3 by eliminating the iterative way of the mean shift procedure. This proposed algorithm shrinks the size of the image by one third of its original image for the computational purpose and then equalizes the number of computations for each new image pixel by constructing links between pixels using their first mean-shift vectors without any iteration process. The accurateness and effectiveness of the proposed methodology is matched with the existing Iterative Mean Shift Algorithm by accomplishing the empherical experiments on the Medical Images (Pathologies Buccales and Eye Retina) composed along with the similarity measures.

BNIMS: Block-based Non-iterative Mean-shift Segmentation algorithm for Medical Images


 
 
 This paper proposed a novel Block based Mean Shift Image Segmentation Algorithm to significantly reduce the computation and improve the segmentation accuracy for high resolution Medical Image. One of the challenging tasks in the image analysis and computer vision area is to correctly classify the pixels as there are no crisp borders among entities in an image. In this proposed methodology, it is observed that the computational complexity of the procedure is diminished by combining the pixels of an image of size MXN into non overlapping image blocks of size 3x3 by eliminating the iterative way of the mean shift procedure. This proposed algorithm shrinks the size of the image by one third of its original image for the computational purpose and then equalizes the number of computations for each new image pixel by constructing links between pixels using their first mean-shift vectors without any iteration process. The accurateness and effectiveness of the proposed methodology is matched with the existing Iterative Mean Shift Algorithm by accomplishing the empherical experiments on the Medical Images (Pathologies Buccales and Eye Retina) composed along with the similarity measures.
 
 

The mean shift algorithm and its relation to kernel regression

We investigate the connection between the asymptotic bias of the well-known Nadaraya–Watson kernel regression and the mean shift (MS) vector with the Gaussian kernel. We first show that the asymptotic bias for the univariate Nadaraya–Watson kernel regression can be estimated using the MS scalar. Then, we investigate the general D-dimensional case (D > 1) and derive a formula for the asymptotic bias as a function of the MS vector. We show that when the regression function is a linear function of its entries, then the asymptotic bias can be represented as a linear function of the MS vector. The MS algorithm for the univariate and the linear cases can be used to find points where the Nadaraya–Watson kernel regression gives an unbiased estimate. Furthermore, this connection suggests that exploiting the theoretical properties of the asymptotic bias of the Nadaraya–Watson kernel regression may be helpful to show the convergence of the MS algorithm. Through the simulations, we show that how the given theoretical results can be used to estimate the bias of the estimated clean signal by just observing the noisy signal. Having access to the bias of the estimator, enables us to evaluate the accuracy of the estimated clean data, which later will be used to design a quantizer.

특징점 추적을 통한 다수 영상의 고속 스티칭 기법

Stitching algorithm obtain a descriptor of the feature points extracted from multiple images, and create a single image through the matching process between the each of the feature points. In this paper, a feature extraction and matching techniques for the creation of a high-speed panorama using video input is proposed. Features from Accelerated Segment Test(FAST) is used for the feature extraction at high speed. A new feature point matching process, different from the conventional method is proposed. In the matching process, by tracking region containing the feature point through the Mean shift vector required for matching is obtained. Obtained vector is used to match the extracted feature points. In order to remove the outlier, the RANdom Sample Consensus(RANSAC) method is used. By obtaining a homography transformation matrix of the two input images, a single panoramic image is generated. Through experimental results, we show that the proposed algorithm improve of speed panoramic image generation compared to than the existing method.