Showing papers on "Centroid published in 2010"

Point Set Registration: Coherent Point Drift

Abstract: Point set registration is a key component in many computer vision tasks. The goal of point set registration is to assign correspondences between two sets of points and to recover the transformation that maps one point set to the other. Multiple factors, including an unknown nonrigid spatial transformation, large dimensionality of point set, noise, and outliers, make the point set registration a challenging problem. We introduce a probabilistic method, called the Coherent Point Drift (CPD) algorithm, for both rigid and nonrigid point set registration. We consider the alignment of two point sets as a probability density estimation problem. We fit the Gaussian mixture model (GMM) centroids (representing the first point set) to the data (the second point set) by maximizing the likelihood. We force the GMM centroids to move coherently as a group to preserve the topological structure of the point sets. In the rigid case, we impose the coherence constraint by reparameterization of GMM centroid locations with rigid parameters and derive a closed form solution of the maximization step of the EM algorithm in arbitrary dimensions. In the nonrigid case, we impose the coherence constraint by regularizing the displacement field and using the variational calculus to derive the optimal transformation. We also introduce a fast algorithm that reduces the method computation complexity to linear. We test the CPD algorithm for both rigid and nonrigid transformations in the presence of noise, outliers, and missing points, where CPD shows accurate results and outperforms current state-of-the-art methods.

Super-resolution Optical Imaging of Single-Molecule SERS Hot Spots

Abstract: We present the first super-resolution optical images of single-molecule surface-enhanced Raman scattering (SM-SERS) hot spots, using super-resolution imaging as a powerful new tool for understanding the interaction between single molecules and nanoparticle hot spots. Using point spread function fitting, we map the centroid position of SM-SERS with ±10 nm resolution, revealing a spatial relationship between the SM-SERS centroid position and the highest SERS intensity. We are also able to measure the unique position of the SM-SERS centroid relative to the centroid associated with nanoparticle photoluminescence, which allows us to speculate on the presence of multiple hot spots within a single diffraction-limited spot. These measurements allow us to follow dynamic movement of the SM-SERS centroid position over time as it samples different locations in space and explores regions larger than the expected size of a SM-SERS hot spot. We have proposed that the movement of the SERS centroid is due to diffusion of ...

Orlicz centroid bodies

Abstract: The sharp affine isoperimetric inequality that bounds the volume of the centroid body of a star body (from below) by the volume of the star body itself is the Busemann-Petty centroid inequality. A decade ago, the $L_p$ analogue of the classical Busemann- Petty centroid inequality was proved. Here, the definition of the centroid body is extended to an Orlicz centroid body of a star body, and the corresponding analogue of the Busemann-Petty centroid inequality is established for convex bodies.

PANORAMA: A 3D Shape Descriptor Based on Panoramic Views for Unsupervised 3D Object Retrieval

Abstract: We present a novel 3D shape descriptor that uses a set of panoramic views of a 3D object which describe the position and orientation of the object's surface in 3D space. We obtain a panoramic view of a 3D object by projecting it to the lateral surface of a cylinder parallel to one of its three principal axes and centered at the centroid of the object. The object is projected to three perpendicular cylinders, each one aligned with one of its principal axes in order to capture the global shape of the object. For each projection we compute the corresponding 2D Discrete Fourier Transform as well as 2D Discrete Wavelet Transform. We further increase the retrieval performance by employing a local (unsupervised) relevance feedback technique that shifts the descriptor of an object closer to its cluster centroid in feature space. The effectiveness of the proposed 3D object retrieval methodology is demonstrated via an extensive consistent evaluation in standard benchmarks that clearly shows better performance against state-of-the-art 3D object retrieval methods.

An active vision system for fall detection and posture recognition in elderly healthcare

Abstract: The paper presents an active vision system for the automatic detection of falls and the recognition of several postures for elderly homecare applications. A wall-mounted Time-Of-Flight camera provides accurate measurements of the acquired scene in all illumination conditions, allowing the reliable detection of critical events. Preliminarily, an off-line calibration procedure estimates the external camera parameters automatically without landmarks, calibration patterns or user intervention. The calibration procedure searches for different planes in the scene selecting the one that accomplishes the floor plane constraints. Subsequently, the moving regions are detected in real-time by applying a Bayesian segmentation to the whole 3D points cloud. The distance of the 3D human centroid from the floor plane is evaluated by using the previously defined calibration parameters and the corresponding trend is used as feature in a thresholding-based clustering for fall detection. The fall detection shows high performances in terms of efficiency and reliability on a large real dataset in which almost one half of events are falls acquired in different conditions. The posture recognition is carried out by using both the 3D human centroid distance from the floor plane and the orientation of the body spine estimated by applying a topological approach to the range images. Experimental results on synthetic data validate the correctness of the proposed posture recognition approach.

Investigation of Spectral Centroid Magnitude and Frequency for Speaker Recognition.

Abstract: Most conventional features used in speaker recognition are based on spectral envelope characterizations such as Mel-scale filterbank cepstrum coefficients (MFCC), Linear Prediction Cepstrum Coefficient (LPCC) and Perceptual Linear Prediction (PLP). The MFCC’s success has seen it become a de facto standard feature for speaker recognition. Alternative features, that convey information other than the average subband energy, have been proposed, such as frequency modulation (FM) and subband spectral centroid features. In this study, we investigate the characterization of subband energy as a two dimensional feature, comprising Spectral Centroid Magnitude (SCM) and Spectral Centroid Frequency (SCF). Empirical experiments carried out on the NIST 2001 and NIST 2006 databases using SCF, SCM and their fusion suggests that the combination of SCM and SCF are somewhat more accurate compared with conventional MFCC, and that both fuse effectively with MFCCs. We also show that frame-averaged FM features are essentially centroid features, and provide an SCF implementation that improves on the speaker recognition performance of both subband spectral centroid and FM features.

Estimation of centroid positions with a matched-filter algorithm: relevance for aberrometry of the eye

Abstract: Most Shack-Hartmann based aberrometers use infrared light, for the comfort of the patients. A large amount of the light that is scattered from the retinal layers is recorded by the detector as background, from which it is not trivial to estimate the centroid of the Shack-Hartmann spot. For a centroiding algorithm, background light can lead to a systematic bias of the centroid positions towards the centre of the software window. We implement a matched filter algorithm for the estimation of the centroid positions of the Shack-Hartmann spots recorded by our aberrometer. We briefly present the performance of our algorithm, and recall the well-known robustness of the matched filter algorithm to background light. Using data collected on 5 human eyes, we parameterise a simple and fast centroiding algorithm and reduce the difference between the two algorithms down to a mean residual wavefront of 0.02 μm rms.

Automatic exposure estimation for HDR images based on image statistics

Abstract: A method of segmenting regions of an image wherein a number of partitions are determined based on a range of an image histogram in a logarithmic luminance domain. Regions are defined by the partitions. A mean value of each region is calculated by K-means clustering wherein the clustering is initialized, data is assigned and centroids are updated. Anchor points are determined based on the centroids and a weight of each pixel is computed based on the anchor points.

Determining Both Surface Position and Orientation in Structured-Light-Based Sensing

Abstract: Position and orientation profiles are two principal descriptions of shape in space. We describe how a structured light system, coupled with the illumination of a pseudorandom pattern and a suitable choice of feature points, can allow not only the position but also the orientation of individual surface elements to be determined independently. Unlike traditional designs which use the centroids of the illuminated pattern elements as the feature points, the proposed design uses the grid points between the pattern elements instead. The grid points have the essences that their positions in the image data are inert to the effect of perspective distortion, their individual extractions are not directly dependent on one another, and the grid points possess strong symmetry that can be exploited for their precise localization in the image data. Most importantly, the grid lines of the illuminated pattern that form the grid points can aid in determining surface normals. In this paper, we describe how each of the grid points can be labeled with a unique color code, what symmetry they possess and how the symmetry can be exploited for their precise localization at subpixel accuracy in the image data, and how 3D orientation in addition to 3D position can be determined at each of them. Both the position and orientation profiles can be determined with only a single pattern illumination and a single image capture.

Centroid-based semi-fragile audio watermarking in hybrid domain

Abstract: Many previous (semi-) fragile audio watermarking schemes adopt binary images as watermarks, which reduce the security of watermarking systems. On the other hand, the content-based or feature-based second generation digital watermarking has limited applicability, and its partial feature points may be damaged by watermarking operations and various common signal processing. To overcome these problems, in this paper, a novel centroid-based semi-fragile audio watermarking scheme in hybrid domain is proposed. The theoretical lower limit of signal to noise ratio (SNR) for objective evaluation of the imperceptibility of watermarked audio signal is deduced, and the watermark embedding capacity and the tamper detection ability are theoretically analyzed. In the proposed scheme, first, the centroid of each audio frame is computed, then Hash function is performed on the obtained centroid to get the watermark, after that, the audio sub-band which carries centroid of audio frame is performed with discrete wavelet transform (DWT) and discrete cosine transform (DCT), and finally the encrypted watermark bits are embedded into the hybrid domain. Theoretical analysis and experimental results show that the proposed scheme is inaudible and applicable to different types of audio signals. Furthermore, its ability of tamper detection and tolerance against common signal processing operations are excellent. Comparing with the existing schemes, the proposed scheme can effectively authenticate the veracity and integrity of audio content, and greatly expand the applicability of the content-based audio watermarking scheme.

Centroid and envelope dynamics of high-intensity charged-particle beams in an external focusing lattice and oscillating wobbler.

Abstract: The centroid and envelope dynamics of a high-intensity charged-particle beam are investigated as a beam smoothing technique to achieve uniform illumination over a suitably chosen region of the target for applications to ion-beam-driven high energy density physics and heavy ion fusion. The motion of the beam centroid projected onto the target follows a smooth pattern to achieve the desired illumination, for improved stability properties during the beam-target interaction. The centroid dynamics is controlled by an oscillating "wobbler," a set of electrically biased plates driven by rf voltage.

Uncertainty analysis for optimum plane extraction from noisy 3D range-sensor point-clouds

Abstract: We utilize a more accurate range noise model for 3D sensors to derive from scratch the expressions for the optimum plane fitting a set of noisy points and for the combined covariance matrix of the plane’s parameters, viz. its normal and its distance to the origin. The range error model used by us is a quadratic function of the true range and also the incidence angle. Closed-form expressions for the Cramer–Rao uncertainty bound are derived and utilized for analyzing four methods of covariance computation: exact maximum likelihood, renormalization, approximate least-squares, and eigenvector perturbation. The effect of the simplifying assumptions inherent in these methods are compared with respect to accuracy, speed, and ease of interpretation of terms. The approximate least-squares covariance matrix is shown to possess a number of desirable properties, e.g., the optimal solution forms its null-space and its components are functions of easily understood terms like the planar-patch’s weighted centroid and scatter. It is also fast to compute and accurate enough in practice. Its experimental application to real-time range-image registration and plane fusion is shown by using a commercially available 3D range sensor.

Systematic error analysis and compensation for high accuracy star centroid estimation of star tracker

Abstract: Subpixel centroid estimation is the most important star image location method of star tracker. This paper presents a theoretical analysis of the systematic error of subpixel centroid estimation algorithm utilizing frequency domain analysis under the consideration of sampling frequency limitation and sampling window limitation. Explicit expression of systematic error of centroid estimation is obtained, and the dependence of systematic error on Gaussian width of star image, actual star centroid location and the number of sampling pixels is derived. A systematic error compensation algorithm for star centroid estimation is proposed based on the result of theoretical analysis. Simulation results show that after compensation, the residual systematic errors of 3-pixel- and 5-pixel-windows’ centroid estimation are less than 2×10−3 pixels and 2×10−4 pixels respectively.

AWCL: Adaptive weighted centroid target localization algorithm based on RSSI in WSN

Abstract: Target localization and tracking is the canonical application of Wireless Sensor Networks. Unlike a centralized system, a sensor network is subject to a unique set of resource constraints such as limited on-board battery power and limited network communication bandwidth. So the traditional tracking algorithm can be directly used in WSN. Therefore efficient localization algorithms that consume less energy for computation and less bandwidth for communication are needed. The weighted centroid localization algorithm (WCL) based on RSSI is applied in most of actual systems. Only one uniform path loss exponent obtained through experiments is used to calculate the weights of nodes in general WCL. It is well known that the path loss exponent is the essential reflection of sensing surroundings. The actual sensing scenario can't be revealed in the traditional WCL algorithm, and therefore it is not appropriate that only one exponent is accepted all through the area covered by the sensor nodes. A new algorithm, adaptively weighted centroid localization (AWCL), is proposed in this paper. Firstly a more reasonable path loss exponent is adaptively estimated according to the surroundings where the target nodes situates. Secondly the target position will be calculated by using the weighted centroid method in which exponents estimated in the first stage are adopted. Theoretical analysis are presented to demonstrate the performance of the proposed localization method, the simulation results show that that the proposed algorithm outperforms the general WCL algorithm.

Centroid Based Detection Algorithm for Hybrid Traffic Sign Recognition System

Abstract: Automatic traffic sign recognition system can help the driver to make a right decision at the right time for safe driving. This paper presents an algorithm for detection of traffic sign using color centroid matching. This algorithm detects the traffic sign from the images captured from the complex road environment. YCbCr color space is used for color segmentation to make the detection process independent of variable illumination characteristic. The proposed method extracts and classifies the detected sign according to colors of the traffic sign. The sign is extracted by considering the maximum distance of boundary pixels from centroid. The sign is further classified into its sub-group according to its shape. The minimum Euclidean distance classifier is used to detect the shape of sign. Perceptron Neural Network (NN) is employed to recognize the classified sign. Results show that the developed algorithm has color classification rate of 100% while shape classification rate about 98% when tested on several outdoor images for traffic sign detection. The overall recognition rate of the developed algorithm is observed around 92%.

Centroid of a general type-2 fuzzy set computed by means of the centroid-flow algorithm

Abstract: The centroid of a general type-2 fuzzy set (T2 FS) A can be obtained by taking the union of the centroids of all the α-planes (each raised to level α) of A. Karnik-Mendel (KM) or the Enhanced Karnik-Mendel (EKM) algorithms are used for computing the centroid of each α-plane. The iterative features in KM/EKM algorithms can be time-consuming, especially when the algorithms have to be repeated for many α-planes. This paper proposes a new method named Centroid Flow (CF) algorithm to compute the centroid of A without having to apply KM/EKM algorithms for every α-plane. Extensive simulations have shown that the CF algorithm can reduce the computation time by 75% to 80% and 50% to 75% compared to KM and EKM algorithms, respectively, and still maintains satisfactory computation accuracy for various T2 FSs when the primary variable x and ar-plane are discretized finely enough.

Minimum variance unbiased subpixel centroid estimation of point image limited by photon shot noise

Abstract: An unbiased subpixel centroid estimation algorithm of point image is proposed through the compensation of the systematic error of the center of mass method. The Cramer-Rao lower bound on centroid estimation variances is derived under the photon shot noise condition and is utilized to evaluate the proposed algorithm. Numerical analysis shows that the proposed centroid estimator attains the required lower bound; thus the proposed algorithm can be asserted as a minimum variance estimator. Simulation results indicate that the centroid accuracy is maximized when the Gaussian width of the signal spot is 0.2-0.3 pixel and the estimator can attain subpixel accuracy close to 1/100 pixel when 1000 photons are detected.

Fast Template Matching Based on Normalized Cross Correlation with Centroid Bounding

Abstract: In this paper, we propose a fast pattern matching algorithm based on normalized cross correlation (NCC) with centroid bounding to achieve very efficient search. The algorithm will calculate histogram around centroid within maximum circle with radius R. After dividing the image into blocks by R×R size, calculating the similarity between the color histograms of the image block and centroid around circle to get potential blocks that the centroid of the template might be in, then by applying NCC to get the final result. Experimental results show the proposed algorithm is very efficient comparing with full-search NCC. The results has broad applications in the fields of object detecting, image retrieval and etc.

Improved Iteratively Weighted Centroiding for accurate spot detection in Laser Guide Star based Shack Hartmann Sensor

Abstract: A hybrid centroiding technique involving Iteratively Weighted Center of Gravity (IWCoG) algorithm and corre-lation technique for a Laser Guide Star (LGS) based Shack Hartmann wavefront sensor is proposed. A simplemethod for simulating LGS elongated spots with photon noise and read out noise is demonstrated. The problemsassociated with IWCoG are addressed (a) Error saturation is minimized by adding random numbers iterativelyto centroid positions, (b) non uniform convergence of Centroid Estimation Error (CEE) is reduced by usingthe hypothesis that the iteration number with maximum correlation between the weighting function and theactual spot image function is the iteration with minimum error, (c) convergence rate is improved by shifting theweighting function to the point of maximum intensity in “rst iteration. The novelty of the algorithm is testedby comparing with other centroiding algorithms.Keywords: Shack Hartmann wavefront sensor, iteratively weighted center of gravity algorithm, adaptive optics,laser guide star

A new method describing border irregularity of pigmented lesions.

Abstract: Background/purpose: Automatic quantitative characterization of border irregularity generating useful descriptors is a highly important task for computer-aided diagnosis of melanoma. This paper proposes a novel approach to describe the border irregularity of melanomas aiming at achieving higher recognition rates. Methods: By introducing a boundary characteristic description, which we call a centroid distance diagram (CDD), a compact-supported mapping, called the centroid distance curve, can be extracted from this diagram. The centroid distance curve establishes the projection from angular orientations to the sum of the lengths of those line segments connecting the lesion centroid and border points. Border irregularity descriptors generated from CDDs include the non-centroid-convexity index, the maximum–minimum distance indicator, the standard deviation of centroid distance curves and the maximum magnitude of non-zero frequency elements of centroid distance curves after discrete Fourier transforms. Upper limits of the error boundaries involved in these descriptors are estimated. Results: Experimental studies are based on 60 melanoma and 107 benign lesion images collected from local pigmented lesion clinics. By applying the proposed descriptors, receiver operating characteristic (ROC) curves are constructed by projecting the features into a linear space learned from samples. The optimal sensitivity and specificity for the proposed method are 74.2% and 72.6%. The total area enclosed by the corresponding ROC curve is 0.788. In addition, as the training and testing study for melanoma recognition in the literature is largely missing, a comprehensive comparative study is conducted by randomly dividing the data into two groups: one for training and one for testing. For the testing group, the best mean sensitivity obtained with the descriptors proposed in this paper reaches 71.8% and the standard deviation is 10.1%. The specificity for the testing group corresponding to the optimal sensitivity is 69.8%, with a standard deviation of 7.2%. Conclusion: This study suggests that in terms of sensitivity, descriptors extracted from CDDs are the most powerful ones in characterizing the border irregularity of melanomas.

Target localization using proximity binary sensors

Abstract: This works presents the maximum likelihood localization (ML) algorithm for multi-target localization using proximity-based sensor networks. Proximity sensors simply report a single binary value indicating whether or not a target is near. The ML approach requires a hill climbing algorithm to find the peak, and its ability to find the global peak is determined by the initial estimates for the target locations. This paper investigates three methods to initialize the ML algorithm: 1) centroid of k-means clustering, 2) centroid of clique clustering, and 3) peak in the 1-target likelihood surface. To provide a performance bound for the initialization methods, the paper also considers the ground truth target positions as initial estimates. Simulations compare the ability of these methods to resolve and localize two targets. The simulations demonstrate that the clique clustering technique out-performs k-means clustering and is nearly as effective as the 1-target likelihood peak methods at a fraction of the computational cost.

Low Cost Artificial Planar Target Measurement Techniques for Terrestrial Laser Scanning

Abstract: SUMMARY This paper presents both an internal and external accuracy assessment of four different methods for measuring the centroid of a signalized planar target captured by a terrestrial laser scanner. The planar targets used in this project are composed of a black background and a white circle printed on 8½ by 11 inches plain sheet of paper using a consumer level LaserJet printer. The first two methods tested define the centroid of a target to be the mean and median of the cluster of points belonging to the white circle in the point cloud. The latter two methods are more advanced, and they take advantage of the planar nature of the target as well as the intensity difference between the circle and the background to strengthen the centroid derivation through a combination of least-squares plane fitting and circle fitting. The main benefit of the four presented methodologies is that no specialized and/or laser scanner dependent targets need to be utilized. And it will be demonstrated in this paper that using the two advanced methods can yield position measurement precision and accuracy far superior to the simple mean or median computations. In fact, sub-millimetre precision and accuracy is achievable from using low cost paper targets provided that an appropriate target measurement algorithm like the latter two methodologies proposed in this paper is adopted.

On the relation between a fuzzy number and its centroid

Abstract: In this paper we answer the following question: if the centroid and the left/right spread of an unknown fuzzy number are given how do we find this fuzzy number?

Adaptive FIR Neural Model for Centroid Learning in Self-Organizing Maps

Abstract: In this paper, a training method for the formation of topology preserving maps is introduced. The proposed approach presents a sequential formulation of the self-organizing map (SOM), which is based on a new model of the neuron, or processing unit. Each neuron acts as a finite impulse response (FIR) system, and the coefficients of the filters are adaptively estimated during the sequential learning process, in order to minimize a distortion measure of the map. The proposed FIR-SOM model deals with static distributions and it computes an ordered set of centroids. Additionally, the FIR-SOM estimates the learning dynamic of each prototype using an adaptive FIR model. A noteworthy result is that the optimized coefficients of the FIR processes tend to represent a moving average filter, regardless of the underlying input distribution. The convergence of the resulting model is analyzed numerically and shows good properties with respect to the classic SOM and other unsupervised neural models. Finally, the optimal FIR coefficients are shown to be useful for visualizing the cluster densities.

Centroid molecular dynamics: Comparison with exact results for model systems

Abstract: The relation between the accuracy of centroid molecular dynamics correlation functions, and the geometry of the centroid potential is investigated. It is shown that, depending on the temperature, there exist several regimes, and in each of them certain features of the exact Kubo correlation functions are reproduced. The change of regimes is related to the emergence of barriers in the centroid potential. In order to clarify how the above described picture of regimes is modified in real systems when dissipation is important, a methodology is developed to test the accuracy of centroid correlation functions for the model of a particle coupled to a harmonic heat bath. A modification of the centroid molecular dynamics method to include the influence of the heat bath is introduced. Preliminary results of comparison of centroid molecular dynamics with the numerically exact results of filtered propagator functional method are presented.

Multiple centroid condensation of probability distribution clouds

Abstract: Systems and methods are disclosed for identifying objects captured by a depth camera by condensing classified image data into centroids of probability that captured objects are correctly identified entities. Output exemplars are processed to detect spatially localized clusters of non-zero probability pixels. For each cluster, a centroid is generated, generally resulting in multiple centroids for each differentiated object. Each centroid may be assigned a confidence value, indicating the likelihood that it corresponds to a true object, based on the size and shape of the cluster, as well as the probabilities of its constituent pixels.

Shack-Hartmann wavefront sensor with large dynamic range.

Abstract: A new spot centroid detection algorithm for a Shack-Hartmann wavefront sensor (SHWFS) is experimentally investigated. The algorithm is a kind of dynamic tracking algorithm that tracks and calculates the corresponding spot centroid of the current spot map based on the spot centroid of the previous spot map, according to the strong correlation of the wavefront slope and the centroid of the corresponding spot between temporally adjacent SHWFS measurements. That is, for adjacent measurements, the spot centroid movement will usually fall within some range. Using the algorithm, the dynamic range of an SHWFS can be expanded by a factor of three in the measurement of tilt aberration compared with the conventional algorithm, more than 1.3 times in the measurement of defocus aberration, and more than 2 times in the measurement of the mixture of spherical aberration plus coma aberration. The algorithm is applied in our SHWFS to measure the distorted wavefront of the human eye. The experimental results of the adaptive optics (AO) system for retina imaging are presented to prove its feasibility for highly aberrated eyes.

Fast algorithm to calculate generalized centroid of interval type-2 fuzzy set

Abstract: The relationship between the switch point on the membership function and the generalized centroid in an interval type-2 fuzzy set is discussed.A close form representation of the switch point for calculating the endpoint of the generalized centroid is provided in both discrete and continuous conditions.Then an opposite direction searching method for computing the switch point is proposed.The convergence of the switch point is proved.Experimental result shows that the approach is computationally faster than the commonly used Enhanced Karnik-Mendel method without loosing numeric precision.