Non-adaptive Sampling Research Articles

Enhanced total variation minimization for stable image reconstruction

The total variation (TV) regularization has phenomenally boosted various variational models for image processing tasks. We propose to combine the backward diffusion process in the earlier literature on image enhancement with the TV regularization, and show that the resulting enhanced TV minimization model is particularly effective for reducing the loss of contrast. The main purpose of this paper is to establish stable reconstruction guarantees for the enhanced TV model from noisy subsampled measurements with two sampling strategies, non-adaptive sampling for general linear measurements and variable-density sampling for Fourier measurements. In particular, under some weaker restricted isometry property conditions, the enhanced TV minimization model is shown to have tighter reconstruction error bounds than various TV-based models for the scenario where the level of noise is significant and the amount of measurements is limited. The advantages of the enhanced TV model are also numerically validated by preliminary experiments on the reconstruction of some synthetic, natural, and medical images.

Adaptive cluster path sampling

Path sampling is one of the economical sampling designs for spatial populations since it provides the capability of sampling all units along a path traveled during a survey. Path sampling is a non-adaptive sampling design; however it is known that the adaptive sampling design provides more precise estimates of the population mean and total than conventional sampling designs. Therefore, the objective of this article is to propose a new sampling design called “adaptive cluster path sampling” using adaptive sampling strategy and conventional path sampling in order to obtain more precise parameter estimate. Two estimators of the population mean and their mean square error were obtained. The efficiency of adaptive cluster path sampling was examined by performing a simulation study, comparing adaptive cluster path sampling to conventional path sampling, adaptive cluster sampling and simple random sampling. According to the simulation results on the blue winged teal data, adaptive cluster path sampling is more efficient than path sampling and adaptive cluster sampling when equal sampling cost is considered.

An adaptive spot placement method on Cartesian grid for pencil beam scanning proton therapy

Pencil beam scanning proton radiotherapy (RT) offers flexible proton spot placement near treatment targets for delivering tumoricidal radiation dose to tumor targets while sparing organs-at-risk. Currently the spot placement is mostly based on a non-adaptive sampling (NS) strategy on a Cartesian grid. However, the spot density or spacing during NS is a constant for the Cartesian grid that is independent of the geometry of tumor targets, and thus can be suboptimal in terms of plan quality (e.g. target dose conformality) and delivery efficiency (e.g. number of spots). This work develops an adaptive sampling (AS) spot placement method on the Cartesian grid that fully accounts for the geometry of tumor targets. Compared with NS, AS places (1) a relatively fine grid of spots at the boundary of tumor targets to account for the geometry of tumor targets and treatment uncertainties (setup and range uncertainty) for improving dose conformality, and (2) a relatively coarse grid of spots in the interior of tumor targets to reduce the number of spots for improving delivery efficiency and robustness to the minimum-minitor-unit (MMU) constraint. The results demonstrate that (1) AS achieved comparable plan quality with NS for regular MMU and substantially improved plan quality from NS for large MMU, using merely about 10% of spots from NS, where AS was derived from the same Cartesian grid as NS; (2) on the other hand, with similar number of spots, AS had better plan quality than NS consistently for regular and large MMU.

Adaptive filter design via a gradient thresholding algorithm for compressive spectral imaging.

Sensing a spectral image data cube has traditionally been a time-consuming task since it requires a scanning process. In contrast, compressive spectral imaging (CSI) has attracted widespread interest since it requires fewer samples than scanning systems to acquire the data cube, thus improving the sensing speed. CSI captures linear projections of the scene, and then a reconstruction algorithm estimates the underlying scene. One notable CSI architecture is the color coded aperture snapshot spectral imager (C-CASSI), which employs pixelated filter arrays as the coding patterns to spatially and spectrally encode the incoming light. Up to date works on C-CASSI have used non-adaptive color coded apertures. Non-adaptive sampling ignores prior information about the signal to design the coding patterns. Therefore, this work proposes a method to adaptively design the color coded aperture, such that the quality of image reconstruction is improved. In more detail, this work introduces a gradient thresholding algorithm, which computes the consecutive color coded aperture from a rapidly reconstructed low-resolution version of the data cube. The successive adaptive patterns enable recovering a data cube in the presence of Gaussian noise with higher image quality. Real reconstructions and simulations evidence an improvement of up to 3dB in the quality of image reconstruction of the proposed method in comparison with state-of-the-art non-adaptive techniques.

Efficient modelling and optimization for double wishbone suspensions based on a non-adaptive sampling sparse response surface

ABSTRACTThe response surface method is always used in optimization problems such as the optimization of vehicle suspensions because of its efficiency. A sparse response surface (SRS) is proposed to represent a double wishbone suspension model constructed from a few non-adaptive sampling points. One set of sampling points can be applied to all response surfaces. In this problem, four variations of the front wheel positioning parameters are considered as the optimization objectives, and four parameters are considered as the design variables. Four SRS models with four variables are constructed from 60 sampling points. The interior point method is employed to optimize the problem. The optimization results show that the objective of the camber angle reduces from to , the objective of the caster angle increases from to , the objective of the kingpin inclination angle reduced from to , and the objective of the toe angle increases from to .

Fuzzy Logic and Regression Approaches for Adaptive Sampling of Multimedia Traffic in Wireless Computer Networks

Organisations such as hospitals and the public are increasingly relying on large computer networks to access information and to communicate multimedia-type data. To assess the effectiveness of these networks, the traffic parameters need to be analysed. Due to the quantity of the data packets, examining each packet’s transmission parameters is not practical, especially in real time. Sampling techniques allow a subset of packets that accurately represents the original traffic to be examined and they are thus important in evaluating the performance of multimedia networks. In this study, an adaptive sampling technique based on regression and a fuzzy inference system was developed. The technique dynamically updates the number of packets sampled by responding to the traffic’s variations. Its performance was found to be superior to the conventional nonadaptive sampling methods.

A novel multi-scale adaptive sampling-based approach for energy saving in leak detection for WSN-based water pipelines

In this paper, we propose a novel monitoring strategy for a wireless sensor networks (WSNs)-based water pipeline network. Our strategy uses a multi-pronged approach to reduce energy consumption based on the use of two types of vibration sensors and pressure sensors, all having different energy levels, and a hierarchical adaptive sampling mechanism to determine the sampling frequency. The sampling rate of the sensors is adjusted according to the bandwidth of the vibration signal being monitored by using a wavelet-based adaptive thresholding scheme that calculates the new sampling frequency for the following cycle. In this multimodal sensing scheme, the duty-cycling approach is used for all sensors to reduce the sampling instances, such that the high-energy, high-precision (HE-HP) vibration sensors have low duty cycles, and the low-energy, low-precision (LE-LP) vibration sensors have high duty cycles. The low duty-cycling (HE-HP) vibration sensor adjusts the sampling frequency of the high duty-cycling (LE-LP) vibration sensor. The simulated test bed considered here consists of a water pipeline network which uses pressure and vibration sensors, with the latter having different energy consumptions and precision levels, at various locations in the network. This is all the more useful for energy conservation for extended monitoring. It is shown that by using the novel features of our proposed scheme, a significant reduction in energy consumption is achieved and the leak is effectively detected by the sensor node that is closest to it. Finally, both the total energy consumed by monitoring as well as the time to detect the leak by a WSN node are computed, and show the superiority of our proposed hierarchical adaptive sampling algorithm over a non-adaptive sampling approach.

Adaptive Sampling Technique Using Regression Modelling and Fuzzy Inference System for Network Traffic.

Electronic-health relies on extensive computer networks to facilitate access and to communicate various types of information in the form of data packets. To examine the effectiveness of these networks, the traffic parameters need to be analysed. Due to quantity of packets, examining their transmission parameters individually is not practical, especially when performed in real time. Sampling allows a subset of packets that accurately represents the original traffic to be chosen. In this study an adaptive sampling method based on regression and fuzzy inference system was developed. It dynamically updates the sampling by responding to the traffic changes. Its performance was found to be superior to the conventional non-adaptive sampling methods.

Rapid real-time generation of super-resolution hyperspectral images through compressive sensing and GPU

ABSTRACTRecently, compressive sensing (CS) has offered a new framework whereby a signal can be recovered from a small number of noisy non-adaptive samples. This is now an active area of research in many image-processing applications, especially super-resolution. CS algorithms are widely known to be computationally expensive. This paper studies a real time super-resolution reconstruction method based on the compressive sampling matching pursuit (CoSaMP) algorithm for hyperspectral images. CoSaMP is an iterative compressive sensing method based on the orthogonal matching pursuit (OMP). Multi-spectral images record enormous volumes of data that are required in practical modern remote-sensing applications. A proposed implementation based on the graphical processing unit (GPU) has been developed for CoSaMP using computed unified device architecture (CUDA) and the cuBLAS library. The CoSaMP algorithm is divided into interdependent parts with respect to complexity and potential for parallelization. The proposed implementation is evaluated in terms of reconstruction error for different state-of-the-art super-resolution methods. Various experiments were conducted using real hyperspectral images collected by Earth Observing-1 (EO-1), and experimental results demonstrate the speeding up of the proposed GPU implementation and compare it to the sequential CPU implementation and state-of-the-art techniques. The speeding up of the GPU-based implementation is up to approximately 70 times faster than the corresponding optimized CPU.

Temporally adaptive sampling: a case study in rare species survey design with marbled salamanders (Ambystoma opacum).

Improving detection rates for elusive species with clumped distributions is often accomplished through adaptive sampling designs. This approach can be extended to include species with temporally variable detection probabilities. By concentrating survey effort in years when the focal species are most abundant or visible, overall detection rates can be improved. This requires either long-term monitoring at a few locations where the species are known to occur or models capable of predicting population trends using climatic and demographic data. For marbled salamanders (Ambystoma opacum) in Massachusetts, we demonstrate that annual variation in detection probability of larvae is regionally correlated. In our data, the difference in survey success between years was far more important than the difference among the three survey methods we employed: diurnal surveys, nocturnal surveys, and dipnet surveys. Based on these data, we simulate future surveys to locate unknown populations under a temporally adaptive sampling framework. In the simulations, when pond dynamics are correlated over the focal region, the temporally adaptive design improved mean survey success by as much as 26% over a non-adaptive sampling design. Employing a temporally adaptive strategy costs very little, is simple, and has the potential to substantially improve the efficient use of scarce conservation funds.

A robust adaptive sampling method for faster acquisition of MR images

A robust adaptive k-space sampling method is proposed for faster acquisition and reconstruction of MR images. In this method, undersampling patterns are generated based on magnitude profile of a fully acquired 2-D k-space data. Images are reconstructed using compressive sampling reconstruction algorithm. Simulation experiments are done to assess the performance of the proposed method under various signal-to-noise ratio (SNR) levels. The performance of the method is better than non-adaptive variable density sampling method when k-space SNR is greater than 10dB. The method is implemented on a fully acquired multi-slice raw k-space data and a quality assurance phantom data. Data reduction of up to 60% is achieved in the multi-slice imaging data and 75% is achieved in the phantom imaging data. The results show that reconstruction accuracy is improved over non-adaptive or conventional variable density sampling method. The proposed sampling method is signal dependent and the estimation of sampling locations is robust to noise. As a result, it eliminates the necessity of mathematical model and parameter tuning to compute k-space sampling patterns as required in non-adaptive sampling methods.

Editoral: Special Issue on Photonic and RF Communications Systems

Editoral: Special Issue on Photonic and RF Communications Systems

Multimedia traffic quality of service management using statistical and artificial intelligence techniques

Managing quality of service (QoS) is an important network operation, especially in hybrid wired and wireless multimedia networks. In this study, a two-stage approach to intelligently manage QoS for multimedia traffic was developed. Voice over Internet protocol (VoIP) was included in the study as an example of a typical multimedia application. Initially an adaptive statistical sampling technique was employed. It determined the traffic's statistics and used them in a fuzzy inference system to determine the optimum interval between every two consecutive sections of the traffic sampled. In the second stage, a fuzzy c-means (FCM) clustering was used to pre-process QoS parameters (delay, jitter and packet loss ratio) obtained from the devised sampling scheme. A multilayer perceptron (MLP) neural network then used the information from FCM to assess the QoS provided for VoIP. It was shown that the developed adaptive statistical sampling represents the traffic more correctly than the systematic, stratified and random non-adaptive sampling methods. Also, the combination of statistical sampling followed by FCMand MLP accurately indicated the QoS for VoIP.

The Sample Complexity of Search Over Multiple Populations

This paper studies the sample complexity of searching over multiple populations. We consider a large number of populations, each corresponding to either distribution P0 or P1. The goal of the search problem studied here is to find one population corresponding to distribution P1 with as few samples as possible. The main contribution is to quantify the number of samples needed to correctly find one such population. We consider two general approaches: non-adaptive sampling methods, which sample each population a predetermined number of times until a population following P1 is found, and adaptive sampling methods, which employ sequential sampling schemes for each population. We first derive a lower bound on the number of samples required by any sampling scheme. We then consider an adaptive procedure consisting of a series of sequential probability ratio tests, and show it comes within a constant factor of the lower bound. We give explicit expressions for this constant when samples of the populations follow Gaussian and Bernoulli distributions. An alternative adaptive scheme is discussed which does not require full knowledge of P1, and comes within a constant factor of the optimal scheme. For comparison, a lower bound on the sampling requirements of any non-adaptive scheme is presented.

Distilled Sensing: Adaptive Sampling for Sparse Detection and Estimation

Adaptive sampling results in significant improvements in the recovery of sparse signals in white Gaussian noise. A sequential adaptive sampling-and-refinement procedure called Distilled Sensing (DS) is proposed and analyzed. DS is a form of multistage experimental design and testing. Because of the adaptive nature of the data collection, DS can detect and localize far weaker signals than possible from non-adaptive measurements. In particular, reliable detection and localization (support estimation) using non-adaptive samples is possible only if the signal amplitudes grow logarithmically with the problem dimension. Here it is shown that using adaptive sampling, reliable detection is possible provided the amplitude exceeds a constant, and localization is possible when the amplitude exceeds any arbitrarily slowly growing function of the dimension.

Group Testing With Probabilistic Tests: Theory, Design and Application

Identification of defective members of large populations has been widely studied in the statistics community under the name of group testing. It involves grouping subsets of items into different pools and detecting defective members based on the set of test results obtained for each pool. In a classical noiseless group testing setup, it is assumed that the sampling procedure is fully known to the reconstruction algorithm, in the sense that the existence of a defective member in a pool results in the test outcome of that pool to be positive. However, this may not be always a valid assumption in some cases of interest. In particular, we consider the case where the defective items in a pool can become independently inactive with a certain probability. Hence, one may obtain a negative test result in a pool despite containing some defective items. As a result, any sampling and reconstruction method should be able to cope with two different types of uncertainty, i.e., the unknown set of defective items and the partially unknown, probabilistic testing procedure. In this work, motivated by the application of detecting infected people in viral epidemics, we design non-adaptive sampling procedures that allow successful identification of the defective items through a set of probabilistic tests. Our design requires only a small number of tests to single out the defective items. In particular, for a population of size $N$ and at most $K$ defective items with activation probability $p$, our results show that $M = O(K^2\log{(N/K)}/p^3)$ tests is sufficient if the sampling procedure should work for all possible sets of defective items, while $M = O(K\log{(N)}/p^3)$ tests is enough to be successful for any single set of defective items. Moreover, we show that the defective members can be recovered using a simple reconstruction algorithm with complexity of $O(MN)$.

Compressive Rendering: A Rendering Application of Compressed Sensing

Recently, there has been growing interest in compressed sensing (CS), the new theory that shows how a small set of linear measurements can be used to reconstruct a signal if it is sparse in a transform domain. Although CS has been applied to many problems in other fields, in computer graphics, it has only been used so far to accelerate the acquisition of light transport. In this paper, we propose a novel application of compressed sensing by using it to accelerate ray-traced rendering in a manner that exploits the sparsity of the final image in the wavelet basis. To do this, we raytrace only a subset of the pixel samples in the spatial domain and use a simple, greedy CS-based algorithm to estimate the wavelet transform of the image during rendering. Since the energy of the image is concentrated more compactly in the wavelet domain, less samples are required for a result of given quality than with conventional spatial-domain rendering. By taking the inverse wavelet transform of the result, we compute an accurate reconstruction of the desired final image. Our results show that our framework can achieve high-quality images with approximately 75 percent of the pixel samples using a nonadaptive sampling scheme. In addition, we also perform better than other algorithms that might be used to fill in the missing pixel data, such as interpolation or inpainting. Furthermore, since the algorithm works in image space, it is completely independent of scene complexity.

Adaptive and nonadaptive approaches to statistically based methods for solving stochastic linear programs: a computational investigation

Large scale stochastic linear programs are typically solved using a combination of mathematical programming techniques and sample-based approximations. Some methods are designed to permit sample sizes to adapt to information obtained during the solution process, while others are not. In this paper, we experimentally examine the relative merits and challenges of approximations based on adaptive samples and those based on non-adaptive samples. We focus our attention on Stochastic Decomposition (SD) as an adaptive technique and Sample Average Approximation (SAA) as a non-adaptive technique. Our results indicate that there can be minimal difference in the quality of the solutions provided by these methods, although comparing their computational requirements would be more challenging.

Minimax Bounds for Active Learning

This paper analyzes the potential advantages and theoretical challenges of algorithms. Active learning involves sequential sampling procedures that use information gleaned from previous samples in order to focus the sampling and accelerate the learning process relative to algorithms, which are based on nonadaptive (usually random) samples. There are a number of empirical and theoretical results suggesting that in certain situations active learning can be significantly more effective than passive learning. However, the fact that active learning algorithms are feedback systems makes their theoretical analysis very challenging. This paper aims to shed light on achievable limits in active learning. Using minimax analysis techniques, we study the achievable rates of classification error convergence for broad classes of distributions characterized by decision regularity and noise conditions. The results clearly indicate the conditions under which one can expect significant gains through active learning. Furthermore, we show that the learning rates derived are tight for boundary fragment classes in d-dimensional feature spaces when the feature marginal density is bounded from above and below.

Estimation following group-sequential response-adaptive clinical trials

A sequential clinical trial model is considered in which two treatments with immediate normally distributed responses are to be compared. The class of one-sided group-sequential tests with response-adaptive sampling developed by Jennison and Turnbull is used to investigate which of the treatments has the larger mean response. The power function for this class of tests is the same as that under nonadaptive sampling, and significant decreases in the inferior treatment number can be achieved with only minor increases in the average total sample number. Two inferential methods are considered following the design. Approximate confidence intervals for the treatment mean difference and the individual means are constructed using the pivotal method of Woodroofe, and an approximation to the bias of the maximum likelihood estimator of the treatment mean difference is studied based on the work of Whitehead. Simulation is used to assess the accuracy of both methods for various stopping boundaries and numbers of interim analyses.