Sparse Localization Research Articles

3D drift correction for super-resolution imaging with a single laser light.

Single-molecule localization microscopy (SMLM) enables three-dimensional (3D) super-resolution imaging of nanoscale structures within biological samples. However, prolonged acquisition introduces a drift between the sample and the imaging system, resulting in artifacts in the reconstructed super-resolution image. Here, we present a novel, to our knowledge, 3D drift correction method that utilizes both the reflected and scattered light from the sample. Our method employs the reflected light of a near-infrared (NIR) laser for focus stabilization while synchronously capturing speckle images to estimate the lateral drift. This approach combines high-precision active compensation in the axial direction with lateral post-processing compensation, achieving the abilities of 3D drift correction with a single laser light. Compared to the popular localization events-based cross correlation method, our approach is much more robust, especially for datasets with sparse localization points.

Multisource Sparse Inversion Localization with Long-Distance Mobile Sensors

To address the threat posed by unknown signal sources within Mobile Crowd Sensing (MCS) systems to system stability and to realize effective localization of unknown sources in long-distance scenarios, this paper proposes a unilateral branch ratio decision algorithm (UBRD). This approach addresses the inadequacies of traditional sparse localization algorithms in long-distance positioning by introducing a time–frequency domain composite block sparse localization model. Given the complexity of localizing unknown sources, a unilateral branch ratio decision scheme is devised. This scheme derives decision thresholds through the statistical characteristics of branch residual ratios, enabling adaptive control over iterative processes and facilitating multisource localization under conditions of remote blind sparsity. Simulation results indicate that the proposed model and algorithm, compared to classic sparse localization schemes, are more suitable for long-distance localization scenarios, demonstrating robust performance in complex situations like blind sparsity, thereby offering broader scenario adaptability.

Joint Sparse Locality Preserving Regression for Discriminative Learning

Ridge Regression (RR) is a classical method that is widely used in multiple regression analysis. However, traditional RR does not take the local geometric structure of data into consideration for discriminative learning and it is sensitive to outliers as it is based on <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$L_{2}$</tex-math></inline-formula> -norm. To address this problem, this article proposes a novel method called Joint Sparse Locality Preserving Regression (JSLPR) for discriminative learning. JSLPR not only applies <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$L_{2,1}$</tex-math></inline-formula> -norm on both loss function and regularization term but also takes the local geometric structure of the data into consideration. The use of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$L_{2,1}$</tex-math></inline-formula> -norm can guarantee the robustness to outliers or noises and the joint sparsity for effective feature selection. Taking the local geometric structure into consideration can improve the performance of the feature extraction and selection method when the data lie on a manifold. To solve the optimization problem of JSLPR, an iterative algorithm is proposed and the convergence of the algorithm is also proven. Experiments on four famous face databases are conducted and the results show the merit of the proposed JSLPR on feature extraction and selection.

InferLoc: Hypothesis-Based Joint Edge Inference and Localization in Sparse Sensor Networks

Ranging-based localization is a fundamental problem in the Internet of Things and unmanned aerial vehicle networks. However, the nodes’ limited-ranging scope and users’ broad coverage purpose inevitably cause network sparsity or subnetwork sparsity. The performances of existing localization algorithms are extremely unsatisfactory in sparse networks. A crucial way to deal with the sparsity is to exploit the hidden knowledge provided by the unmeasured edges, which inspires this work to propose a hypothesis-based Joint Edge Inference and Localization algorithm called InferLoc . InferLoc mines the Unmeasured but Inferable Edges (UIEs). Each UIE is an unmeasured edge, but it is restricted through other edges in the network to be inside a rigid component, so it has only a limited number of possible lengths. We propose an efficient method to detect UIEs and geometric approaches to infer possible lengths for UIEs in 2D and 3D networks. The inferred possible lengths of UIEs are then treated as multiple hypotheses to determine the node locations and the lengths of UIEs simultaneously through a joint graph optimization process. In the joint graph optimization model, to make the 0/1 decision variables for hypotheses selection differentiable, differentiable functions are proposed to relax the 0/1 selections, and rounding is applied to select the final length after the optimization converges. We also prove the condition when a UIE can contribute to sparse localization. Extensive experiments show remarkably better accuracy and efficiency performances of InferLoc than the state-of-the-art network localization algorithms. In particular, it reduces the localization errors by more than 90% and speeds up the convergence time more than 100 times than that of the widely used G2O-based methods in sparse networks.

3D Off-Grid Localization for Adjacent Cavitation Noise Sources Using Bayesian Inference

The propeller tip vortex cavitation (TVC) localization problem involves the separation of noise sources in proximity. This work describes a sparse localization method for off-grid cavitations to estimates their precise locations while keeping reasonable computational efficiency. It adopts two different grid (pairwise off-grid) sets with a moderate grid interval and provides redundant representations for adjacent noise sources. To estimate the position of the off-grid cavitations, a block-sparse Bayesian learning-based method is adopted for the pairwise off-grid scheme (pairwise off-grid BSBL), which iteratively updates the grid points using Bayesian inference. Subsequently, simulation and experimental results demonstrate that the proposed method achieves the separation of adjacent off-grid cavitations with reduced computational cost, while the other scheme suffers from a heavy computational burden; for the separation of adjacent off-grid cavitations, the pairwise off-grid BSBL took significantly less time (29 s) compared with the time taken by the conventional off-grid BSBL (2923 s).

Three-dimensional wide-field fluorescence microscopy for transcranial mapping of cortical microcirculation

Wide-field fluorescence imaging is an indispensable tool for studying large-scale biodynamics. Limited space-bandwidth product and strong light diffusion make conventional implementations incapable of high-resolution mapping of fluorescence biodistribution in three dimensions. We introduce a volumetric wide-field fluorescence microscopy based on optical astigmatism combined with fluorescence source localization, covering 5.6×5.6×0.6 mm3 imaging volume. Two alternative configurations are proposed exploiting multifocal illumination or sparse localization of point emitters, which are herein seamlessly integrated in one system. We demonstrate real-time volumetric mapping of the murine cortical microcirculation at capillary resolution without employing cranial windows, thus simultaneously delivering quantitative perfusion information across both brain hemispheres. Morphological and functional changes of cerebral vascular networks are further investigated after an acute ischemic stroke, enabling cortex-wide observation of concurrent collateral recruitment events occurring on a sub-second scale. The reported technique thus offers a wealth of unmatched possibilities for non- or minimally invasive imaging of biodynamics across scales.

Dynamic distribution modelling of the swamp tigertail dragonflySynthemis eustalacta(Odonata: Anisoptera: Synthemistidae) over a 20‐year bushfire regime

AbstractIntensity and severity of bushfires in Australia have increased over the past few decades due to climate change, threatening habitat loss for numerous species. Although the impact of bushfires on vertebrates is well‐documented, the corresponding effects on insect taxa are rarely examined, although they are responsible for key ecosystem functions and services. Understanding the effects of bushfire seasons on insect distributions could elucidate long‐term impacts and patterns of ecosystem recovery.Here, the authors investigated the effects of recent bushfires, land‐cover change, and climatic variables on the distribution of a common and endemic dragonfly, the swamp tigertail (Synthemis eustalacta) (Burmeister, 1839), which inhabits forests that have recently undergone severe burning. The authors used a temporally dynamic species distribution modelling approach that incorporated 20 years of community‐science data on dragonfly occurrence and predictors based on fire, land cover, and climate to make yearly predictions of suitability. The authors also compared this to an approach that combines multiple temporally static models that use annual data.The authors found that for both approaches, fire‐specific variables had negligible importance for the models, while the percentage of tree and non‐vegetative cover were most important. The authors also found that the dynamic model outperformed the static ones, based on cross‐validation omission rate. Model predictions indicated temporal variation in area and spatial arrangement of suitable habitat, but no patterns of habitat expansion, contraction, or shifting.These results highlight not only the efficacy of dynamic modelling to capture spatiotemporal variables such as vegetation cover for an endemic insect species, but also provide a novel approach to mapping species distributions with sparse locality records.

On the localization of oscillatory sources from (S)EEG recordings.

Electrophysiological brain source localization consists in estimating the source positions and activities responsible for the (S)EEG measurements. The localization procedure is usually carried out in the time domain, however in specific situations the activities of interest can be located at well defined frequencies, e.g. in response to a rhythmic stimulation. This paper addresses the problem of sparse localization of multiple sources oscillating at the same frequency. In particular the non-unicity of the solution is emphasized, as alternative source maps involving equivalent or less number of sources can be found, challenging source localization methods based on sparsity. These limitations are illustrated under a realistic SEEG simulation framework, and the usefulness to perform localization for this modality is strengthen out.

Constrained L1-Norm Minimization Method for Range-Based Source Localization under Mixed Sparse LOS/NLOS Environments.

Under mixed sparse line-of-sight/non-line-of-sight (LOS/NLOS) conditions, how to quickly achieve high positioning accuracy is still a challenging task and a critical problem in the last dozen years. To settle this problem, we propose a constrained L1 norm minimization method which can reduce the effects of NLOS bias for improve positioning accuracy and speed up calculation via an iterative method. We can transform the TOA-based positioning problem into a sparse optimization one under mixed sparse LOS/NLOS conditions if we consider NLOS bias as outliers. Thus, a relatively good method to deal with sparse localization problem is L1 norm. Compared with some existing methods, the proposed method not only has the advantages of simple and intuitive principle, but also can neglect NLOS status and corresponding NLOS errors. Experimental results show that our algorithm performs well in terms of computational time and positioning accuracy.

Dimensionality Reduction with Sparse Locality for Principal Component Analysis

Various dimensionality reduction (DR) schemes have been developed for projecting high-dimensional data into low-dimensional representation. The existing schemes usually preserve either only the global structure or local structure of the original data, but not both. To resolve this issue, a scheme called sparse locality for principal component analysis (SLPCA) is proposed. In order to effectively consider the trade-off between the complexity and efficiency, a robust L2,p-norm-based principal component analysis (R2P-PCA) is introduced for global DR, while sparse representation-based locality preserving projection (SR-LPP) is used for local DR. Sparse representation is also employed to construct the weighted matrix of the samples. Being parameter-free, this allows the construction of an intrinsic graph more robust against the noise. In addition, simultaneous learning of projection matrix and sparse similarity matrix is possible. Experimental results demonstrate that the proposed scheme consistently outperforms the existing schemes in terms of clustering accuracy and data reconstruction error.

Where did the tumor start? An inverse solver with sparse localization for tumor growth models

We present a numerical scheme for solving an inverse problem for parameter estimation in tumor growth models for glioblastomas, a form of aggressive primary brain tumor. The growth model is a reaction–diffusion partial differential equation (PDE) for the tumor concentration. We use a PDE-constrained optimization formulation for the inverse problem. The unknown parameters are the reaction coefficient (proliferation), the diffusion coefficient (infiltration), and the initial condition field for the tumor PDE. Segmentation of magnetic resonance imaging (MRI) scans drive the inverse problem where segmented tumor regions serve as partial observations of the tumor concentration. Like most cases in clinical practice, we use data from a single time snapshot. Moreover, the precise time relative to the initiation of the tumor is unknown, which poses an additional difficulty for inversion. We perform a frozen-coefficient spectral analysis and show that the inverse problem is severely ill-posed. We introduce a biophysically motivated regularization on the structure and magnitude of the tumor initial condition. In particular, we assume that the tumor starts at a few locations (enforced with a sparsity constraint on the initial condition of the tumor) and that the initial condition magnitude in the maximum norm is equal to one. We solve the resulting optimization problem using an inexact quasi-Newton method combined with a compressive sampling algorithm for the sparsity constraint. Our implementation uses PETSc and AccFFT libraries. We conduct numerical experiments on synthetic and clinical images to highlight the improved performance of our solver over a previously existing solver that uses standard two-norm regularization for the calibration parameters. The existing solver is unable to localize the initial condition. Our new solver can localize the initial condition and recover infiltration and proliferation. In clinical datasets (for which the ground truth is unknown), our solver results in qualitatively different solutions compared to the two-norm regularized solver.

Jointly Sparse Locality Regression for Image Feature Extraction

This paper proposes a novel method called Jointly Sparse Locality Regression (JSLR) for feature extraction and selection. JSLR utilizes joint $L_{2,1}$ -norm minimization on regularization term, and also introduces the locality to characterize the local geometric structure of the data. There are three main contributions in JSLR for face recognition. Firstly, it eliminates the drawback in ridge regression and Linear Discriminant Analysis (LDA) that when the number of the classes is too small, not enough projections can be obtained for feature extraction. Secondly, by using the local geometric structure as the regularization term, JSLR is able to preserve local information and find an embedding subspace which can detect the most essential data manifold structure. Moreover, since the $L_{2,1}$ -norm based loss function is robust to outliers in data points, JSLR provides the joint sparsity for robust feature selection. The theoretical connections of the proposed method and the previous regression methods are explored and the convergence of the proposed algorithm is also proved. Experimental evaluation on several well-known data sets shows the merits of the proposed method on feature selection and classification.

Semi-supervised dimensionality reduction via sparse locality preserving projection

The dimensionality reduction of the unbalanced semi-supervised problem is difficult because there are too few labeled samples. In this paper, we propose a new dimensionality reduction method for the unbalanced semi-supervised problem, called sparse locality preserving projection (SLPP for short). In the past work of solving the semi-supervised dimensionality reduction problems, they either abandon some unlabeled samples or do not utilize the implicit discriminant information of unlabeled samples. While, SLPP learns the optimal projection matrix with the full use of the discriminant information and the geometric structure of the unlabeled samples. Here, we preserve the geometric structure of the rest unlabeled samples and their k-nearest neighbors after increasing the number of labeled samples by label propagation. The optimization problem of SLPP can be easily solved by a generalized eigenvalue problem. Results on various data sets from UCI machine learning repository and two hyperspectral data sets demonstrate that SLPP is superior to other conventional reduction methods.

Scale-Aware Feature Network for Weakly Supervised Semantic Segmentation

Weakly supervised semantic segmentation with image-level labels is of great significance since it alleviates the dependency on dense annotations. However, as it relies on image classification networks that are only capable of producing sparse object localization maps, its performance is far behind that of fully supervised semantic segmentation models. Inspired by the successful use of multi-scale features for an improved performance in a wide range of visual tasks, we propose a Scale-Aware Feature Network (SAFN) for generating object localization maps. The proposed SAFN uses an attention module to learn the relative weights of multi-scale features in a modified fully convolutional network with dilated convolutions. This approach leads to efficient enlargements of the receptive fields of view and produces dense object localization maps. Our approach achieves mIoUs of 62.3% and 66.5% on the PASCAL VOC 2012 test set using VGG16 based and ResNet based segmentation models, respectively, outperforming other state-of-the-art methods for the weakly supervised semantic segmentation task.

A Node Self-Localization Algorithm With a Mobile Anchor Node in Underwater Acoustic Sensor Networks

Aiming at improving the localization accuracy, reducing network cost, and energy loss in the 3-D environment, this paper proposes a mobile node localization algorithm based on compressive sensing for underwater acoustic sensor networks (UASNs). By dividing the cubic module of the underwater monitoring area and using the energy between the mobile anchor node and the unknown nodes, the sparse localization problem based on the cubic module is transformed into the nodes localization problem based on compressive sensing. Since the energy localization between nodes based on cubic modularization is adopted, the path of the mobile node does not need to be specially designed, as long as the mobile beacon node traverses the entire cubic module. Considering the distance problem of the moving path, the mobile node path is based on the random waypoint (RWP) and the LAYERED-SCAN model. The simulation results show that the proposed algorithm can be applied to the node localization problem of UASNs. It can reduce network cost and node energy loss while obtaining higher localization accuracy.

Sparse locality preserving discriminative projections for face recognition

Recently, the construction of intrinsic graph using sparse representation (SR) has attracted considerable interest. Comparing with the traditional construction methods like k-NN and ε-ball which can well preserve the manifold structure of samples, SR method is more robust to data noise and parameter-free. To exploit the merits of robustness of sparse representation and manifold learning, we propose a new algorithm called sparse locality preserving discriminative projections (SLPDP), which utilizes sparse representation to construct the intrinsic weighted matrix of training samples and incorporates “locality” and “sparsity” into objective function. Simultaneously, SLPDP takes into account the global information of samples like LDPD and DSNPE, and integrates maximum margin criterion (MMC) into the optimal functions for dimensionality reduction. Experiments on PIE, AR, Extended Yale B and Yale face image databases demonstrate the effectiveness of the proposed approach.

Wavefield separation projector processing with a prioriknowledge for acoustic source localization in rooms

In bounded media, wavefield separation projector is an array processing used to extract the direct path from the reverberated ones. Associated with a simple sparse localization algorithm, it makes an attractive method for source localization in rooms of unknown shape. However, the large number of microphones required by the method can prevent its applicability. In this work, we propose to add three different a priori knowledge to dramatically decrease the number of microphones. First, an estimation of the critical distance allows to reduce the required rank of the projection operator and therefore the number of receivers. The second method increases the number of receivers thanks to “virtual measurements” on the boundaries, when the room geometry is partially known. Finally, the last method requires a simple calibration step based on the passive recovering of the Green’s functions between all the pairs of microphones, which also extends the model to weakly inhomogeneous propagation medium. The properties of the 3 methods are discussed. We show numerically and experimentally that these methods lead to a precise source localization, with a moderate number of microphones.

High Accuracy Human Activity Recognition Based on Sparse Locality Preserving Projections.

Human activity recognition(HAR) from the temporal streams of sensory data has been applied to many fields, such as healthcare services, intelligent environments and cyber security. However, the classification accuracy of most existed methods is not enough in some applications, especially for healthcare services. In order to improving accuracy, it is necessary to develop a novel method which will take full account of the intrinsic sequential characteristics for time-series sensory data. Moreover, each human activity may has correlated feature relationship at different levels. Therefore, in this paper, we propose a three-stage continuous hidden Markov model (TSCHMM) approach to recognize human activities. The proposed method contains coarse, fine and accurate classification. The feature reduction is an important step in classification processing. In this paper, sparse locality preserving projections (SpLPP) is exploited to determine the optimal feature subsets for accurate classification of the stationary-activity data. It can extract more discriminative activities features from the sensor data compared with locality preserving projections. Furthermore, all of the gyro-based features are used for accurate classification of the moving data. Compared with other methods, our method uses significantly less number of features, and the over-all accuracy has been obviously improved.

Broadband underwater multi-source localization with a computationally efficient coherent OMP algorithm

In this paper, we propose a broadband coherent matched-field processing (MFP) algorithm to solve multi-source localization problems in shallow water scenarios. The proposed algorithm combines the matched-phase coherent processor with the sparse recovery technique from compressive sensing (CS) theory. A greedy sparse recovery algorithm is adopted to iteratively locate multiple sources using a matched-phase coherent processor. At each iteration of the greedy algorithm, the data is processed coherently using the phase descent search (PDS) algorithm, rather than the incoherent methods used in many sparse recovery algorithms, such as the classical orthogonal matching pursuit (OMP) algorithm. The phase shifts between different frequencies are estimated and compensated, such that the performance can be greatly enhanced. The proposed algorithm is applied to simulated data, synthesized data, and data collected in the SWellEx-96 shallow water experiment. The result provides sparse localization information that matches the ground truth source locations in the simulation and the source trajectory calculated from the Global Positioning System (GPS) information from the experiment.

A Multiple Target Localization with Sparse Information in Wireless Sensor Networks

It is a great challenge for wireless sensor network to provide enough information for targets localization due to the limits on application environment and its nature, such as energy, communication, and sensing precision. In this paper, a multiple targets localization algorithm with sparse information (MTLSI) was proposed using compressive sensing theory, which can provide targets position with incomplete or sparse localization information. It does not depend on extra hardware measurements. Only targets number detected by sensors is needed in the algorithm. The monitoring region was divided into a plurality of small grids. Sensors and targets are randomly dropped in grids. Targets position information is defined as a sparse vector; the number of targets detected by sensor nodes is expressed as the product of measurement matrix, sparse matrix, and sparse vector in compressive sensing theory. Targets are localized with the sparse signal reconstruction. In order to investigate MTLSI performance, BP and OMP are applied to recover targets localization. Simulation results show that MTLSI can provide satisfied targets localization in wireless sensor networks application with less data bits transmission compared to multiple targets localization using compressive sensing based on received signal strengths (MTLCS-RSS), which has the same computation complexity as MTLIS.