Simultaneous tracking and classification: a modularized scheme

Methods dealing with the problem of Joint Tracking and Classification (JTC) are abundant, among which Simultaneous Tracking and Classification (STC) provides a modularized scheme solving tracking and classification subproblems simultaneously. However, there is no explicit hard decision on the class label but only soft decision (class probability) is provided. This does not fit many practical cases, in which a hard decision is urgently needed. To solve this problem, this paper proposes a Hard decision-based STC (HSTC) method. HSTC takes all the decision error rate, timeliness, and estimation error into account. Specifically, for decision, the sequential probability ratio test is adopted due to its nice properties and also the adaptability to our situation. For estimation, by utilizing the two-way information exchange between the tracker and the classifier, we propose flexible three tracking schemes related to decision. The HSTC tracking result is divided into three parts according to the time of making the hard decision. In general, the proposed HSTC method takes advantage of both SPRT and STC. Finally, two illustrative JTC examples with hard decision verify the effectiveness of the the proposed HSTC method. They show that HSTC can meet the demand of the problem, and also has the performance superiority in both decision and estimation.

This paper addresses the problem of joint tracking and classification (JTC) of a single extended target with a complex shape. To describe this complex shape, the spatial extent state is first modeled by star-convex shape via a random hypersurface model (RHM), and then used as feature information for target classification. The target state is modeled by two vectors to alleviate the influence of the high-dimensional state space and the severely nonlinear observation model on target state estimation, while the Euclidean distance metric of the normalized Fourier descriptors is applied to obtain the analytical solution of the updated class probability. Consequently, the resulting method is called the “JTC-RHM method.” Besides, the proposed JTC-RHM is integrated into a Bernoulli filter framework to solve the JTC of a single extended target in the presence of detection uncertainty and clutter, resulting in a JTC-RHM-Ber filter. Specifically, the recursive expressions of this filter are derived. Simulations indicate that: (1) the proposed JTC-RHM method can classify the targets with complex shapes and similar sizes more correctly, compared with the JTC method based on the random matrix model; (2) the proposed method performs better in target state estimation than the star-convex RHM based extended target tracking method; (3) the proposed JTC-RHM-Ber filter has a promising performance in state detection and estimation, and can achieve target classification correctly.

In this paper, single-target tracking using radar measurements is addressed. Recently, algorithms based on Bernoulli random finite sets have proved efficient in a cluttered environment. However, in Bayesian approaches, the choice of the motion model impacts the trajectory estimation accuracy. To select an appropriate set of motion models, a joint tracking and classification (JTC) algorithm can be used. The principle is to consider different target classes depending on their maneuvrability, each of them being associated to a set of motion models. In this context, additional information such as a target length extent measurement can improve both classification and trajectory estimation. Therefore, we propose a multiple-model Bernoulli filter to perform JTC. To jointly estimate the trajectory and the target length which is constant, a Rao-Blackwellized approach is considered. Another contribution is that a bank of probabilistic data association filters is run instead of Kalman filters to account for false detections.

This paper presents a new method for frame rate up‐conversion using sub‐sampled bilateral motion estimation, motion vector smoothing, simplified bidirectional motion compensated interpolation, and complex motion detection. The proposed method produces improved image quality and requires less computational complexity compared to conventional motion estimation. The average PSNRs of the proposed method for frame rate up‐conversion outperform those of the benchmark frame rate up‐conversion technique by up to 4.39 dB for the test sequences. Furthermore, the motion estimation of the proposed FRUC method reduces the required computational complexity to 12.5% of that required for full‐search block matching motion estimation.

This paper deals with the problem of coordinating a team of mobile sensor platforms searching for a single mobile non-evading target. It follows the general Bayesian active sensor network approach introduced in [2] where each decision maker plans locally based on an equivalent representation of the target state probability density function (PDF). This paper focuses on the prediction stage of the decentralized Bayesian filter. It looks at how different types of realistic external constraints may affect the target motion and how they may be taken into account in the process model. Two general classes of constraints are identified soft and hard. A few constraint examples from each class are given to illustrate their impact on the evolution of the target state PDF. Multiple constraints of various types can be combined to increase the accuracy of the predicted PDF estimate, thus affecting the individual trajectories of the search platforms. The effectiveness of the framework is demonstrated for a team of airborne search vehicles looking for a drifting target lost in a storm at sea.

Aiming at problems that the eigenvalue based spectrum sensing algorithms don’t perform well in the situation of low SNR, small sample and need high computational complexity with eigenvalue decomposition, based on the difference value between maximum and minimum eigenvalue spectrum sensing algorithm (DMM), a difference value between the arithmetic mean and geometric mean eigenvalue spectrum sensing algorithm (DAG) with low computational complexity and dynamic threshold was proposed, which via the power method. Simulation results show that the DAG can improve performance over the classical algorithms in situation of low SNR, small samples and increased second users without reduction of computational complexity.

BackgroundPhysical exercise is an important method for both the physical and mental health of the senior population. However, excessive exertion can lead to increased risks of falls, severe injuries, and diminished quality of life. Therefore, simple and effective methods for fatigue monitoring during exercise are highly desirable, particularly in community settings. The purpose of this study was to explore the possibility of real-time detection of exercise-induced fatigue using surface Electromyogram (sEMG) features, including the kurtosis and skewness of the Probability Density Function (PDF) in the community settings to solve the issues of low sensitivity and high computational complexity of commonly used sEMG features.MethodssEMG signals from six forearm muscles were recorded during hand grip tasks at 20% maximal voluntary contraction (MVC) task-to-failure contractions from 30 healthy community-dwelling elders at their respective community centers. PDF shape features of the sEMG, namely kurtosis and skewness, were computed from 25 s of non-fatigue stable phase and 25 s of fatigue data for comparison. Statistical tests were conducted to compare and test for the significance of these features. We further proposed a novel fatigue indicator, Temporal-Mean-Kurtosis (TMK) of channel-averaged kurtosis, to detect fatigue with relatively low computational complexity and adequate sensitivity in community settings. ANOVA and post-hoc analyses were performed to examine the performance of TMK.ResultsStatistically significant differences were found between the non-fatigue period and the fatigue period for both kurtosis and skewness, with increasing values when approaching fatigue. TMK was shown to be sensitive in detecting fatigue with respect to time with lower computational complexity than the Sample Entropy.ConclusionThis study investigated PDF shape features of sEMG signals during a handgrip exercise to identify muscle fatigue in older adults in community experiments. Results revealed significant changes in kurtosis upon fatigue, indicating that PDF shape features were suitable convenient detectors of muscle fatigue in community experiments. The proposed indicator, TMK, showed potential sensitivity in tracking muscle fatigue over time in community-based settings with limited computational complexity, highlighting the promise of sEMG’s PDF features in detecting muscle fatigue among the elderly.

This article aims at performing maritime target classification in SAR images using machine learning (ML) and deep learning (DL) techniques. In particular, the targets of interest are oil platforms and ships located in the Campos Basin, Brazil. Two convolutional neural networks (CNNs), VGG-16 and VGG-19, were used for attribute extraction. The logistic regression (LR), random forest (RF), support vector machine (SVM), k-nearest neighbours (kNN), decision tree (DT), naive Bayes (NB), neural networks (NET), and AdaBoost (ADBST) schemes were considered for classification. The target classification methods were evaluated using polarimetric images obtained from the C-band synthetic aperture radar (SAR) system Sentinel-1. Classifiers are assessed by the accuracy indicator. The LR, SVM, NET, and stacking results indicate better performance, with accuracy ranging from 84.1% to 85.5%. The Kruskal–Wallis test shows a significant difference with the tested classifier, indicating that some classifiers present different accuracy results. The optimizations provide results with more significant accuracy gains, making them competitive with those shown in the literature. There is no exact combination of methods for SAR image classification that will always guarantee the best accuracy. The optimizations performed in this article were for the specific data set of the Campos Basin, and results may change depending on the data set format and the number of images.

Modular soft robot arms (MSRAs) are composed of multiple modules connected in a sequence, and they can bend at different angles in various directions. This capability allows MSRAs to perform more intricate tasks than single-module robots. However, the modular structure also induces challenges in accurate planning and control. Nonlinearity and hysteresis complicate the physical model, while the modular structure and increased DOFs further lead to cumulative errors along the sequence. To address these challenges, we propose a versatile configuration space planning and control strategy for MSRAs, named <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$S2C2A$</tex-math></inline-formula> (State to Configuration to Action). Our approach formulates an optimization problem, <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$S2C$</tex-math></inline-formula> (State to Configuration planning), which integrates various loss functions and a forward model based on biLSTM to generate configuration trajectories based on target states. A configuration controller <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$C2A$</tex-math></inline-formula> (Configuration to Action control) based on biLSTM is implemented to follow the planned configuration trajectories, leveraging only inaccurate internal sensing feedback. We validate our strategy using a cable-driven MSRA, demonstrating its ability to perform diverse offline tasks such as position and orientation control and obstacle avoidance. Furthermore, our strategy endows MSRA with online interaction capability with targets and obstacles. Future work focuses on addressing MSRA challenges, such as more accurate physical models.

The computational complexity and system bit-error-rate (BER) performance of four types of neural-network-based nonlinear equalizers are analyzed for a 50-Gb/s pulse amplitude modulation (PAM)-4 direct-detection (DD) optical link. The four types are feedforward neural networks (F-NN), radial basis function neural networks (RBF-NN), auto-regressive recurrent neural networks (AR-RNN) and layer-recurrent neural networks (L-RNN). Numerical results show that, for a fixed BER threshold, the AR-RNN-based equalizers have the lowest computational complexity. Amongst all the nonlinear NN-based equalizers with the same number of inputs and hidden neurons, F-NN-based equalizers have the lowest computational complexity while the AR-RNN-based equalizers exhibit the best BER performance. Compared with F-NN or RNN, RBF-NN tends to require more hidden neurons with the increase of the number of inputs, making it not suitable for long fiber transmission distance. We also demonstrate that only a few tens of multiplications per symbol are needed for NN-based equalizers to guarantee a good BER performance. This relatively low computational complexity signifies that various NN-based equalizers can be potentially implemented in real time. More broadly, this paper provides guidelines for selecting a suitable NN-based equalizer based on BER and computational complexity requirements.

Endmember extraction is very important in hyperspectral image analysis. The accurate identification of endmembers enables target detection and classification and efficient spectral unmixing. Although a number of endmember extraction algorithms have been proposed, such as two state-of-the-art algorithms—vertex component analysis (VCA) and simplex growing algorithm (SGA)—it is still a rather challenging task. In this paper, a new maximum simplex volume method based on Householder transformation (HT), referred to as maximum volume by HT (MVHT), is presented for endmember extraction. The proposed algorithm provides consistent results with low computational complexity, which overcomes the disadvantage of the inconsistent result of VCA and the shortcoming of the high computational cost of SGA resulted from calculating the simplex volume. A comparative study and analysis are conducted among the three endmember extraction algorithms, VCA, SGA, and MVHT, on both simulated and real hyperspectral data. The obtained experimental results demonstrate that the proposed MVHT algorithm generally provides a competitive or even better performance over VCA and SGA.

Owing to low cost and relatively accurate range measurement, ultrasonic sensors are widely used in various simultaneous localisation and mapping (SLAM) applications. In spite of the abundance of ultrasonic sensor based SLAM applications, a simple and efficient algorithm for an ultrasonic sensor based positioning system with good accuracy and low computational complexity has not yet emerged. The major difficulty is the trade-off between localisation accuracy and computational complexity in most SLAM algorithms, such as extended Kalman filter (EKF) and particle filter. Typically, they improve localisation accuracy by increasing the density of the landmarks, as a result leading to high computational complexity of algorithms and limiting the utilisation of algorithms into online SLAM systems. This study addresses an improved particle filter algorithm to solve ultrasonic sensor based 2D range-only SLAM problem with relatively good accuracy and low computational complexity. This algorithm uses a simple four fixed features based system model to limit the density of the landmarks. A technique called map adjustment is proposed to increase the accuracy and efficiency of the algorithm. Using map adjustment, the proposed particle filter algorithm can improve localisation accuracy and lower computational complexity. The experiment results demonstrate that this algorithm has a better performance than conventional particle filter localisation algorithm.

A low computational complexity bandwidth extension method for mobile audio coding

Millimeter-wave radar is widely used in intelligent vehicle infrastructure cooperative systems (IVICS) because of its long detection distance, high integration and less affected by weather conditions. The target recognition and classification technology based on millimeter wave radar is a research hotspot. However, traditional recognition methods require a complex feature extraction process. In this paper, a millimeter-wave radar classification method based on deep learning is proposed, which uses the ability of convolutional neural networks (CNN) method to automatically extract feature data, so as to replace most of the complex processes of traditional radar signal processing chain. It solves the four classification tasks of bus, cars, electromobile and pedestrians. Taking the selected feature data as the RV spectrum, a seven layer convolution neural network is built to classify the target. Compared with traditional machine learning methods, CNN greatly reduces the workload of feature engineering, and its classification effect is better than traditional machine learning methods. Then, from the perspective of redundancy of the convolution kernels, we simplify the CNN architecture. As a result, final classification accuracy of the CNN is 93.16%. After simplifying the structure of the CNN, the computational complexity and storage consumption of the network are reduced by 74.67% and 73.71%, respectively, while ensuring that the classification accuracy does not decline.

A novel sign adaptation scheme for convex combination of two adaptive filters