Conventional Single Sensor Research Articles

Online Multipedestrian Tracking Based on Fused Detections of Millimeter Wave Radar and Vision

This paper focuses on the challenge of pedestrian tracking using camera and millimeter wave radar in autonomous driving. Pedestrian tracking using a single sensor has inherent limitations due to the lack of comprehensive dimensionality of tracking information. Meanwhile, existing multi-sensor based tracking algorithms suffer from limited tracking accuracy by applying the fusion of projected positions. To enhance the tracking accuracy and robustness, a multi-sensor based tracking algorithm based on fused detection of millimeter-wave radar and vision is proposed, which improves the association of detection results from multiple heterogeneous sensors by utilizing newly designed back-projection and undirected graph, and finally improves the fusion detection by simultaneously utilizing a pedestrian’s appearance, local and global location information. Field tests are conducted to produce dataset, and the performance evaluation results based on the self-produced dataset have verified the superiority of the proposed algorithm over the conventional single-sensor based tracking algorithm and multi-sensor based tracking algorithms.

Object Detection Routine for Material Streams Combining RGB and Hyperspectral Reflectance Data Based on Guided Object Localization

Electronic waste is the fastest growing type of scrap globally and is an important challenge due to its heterogeneity, intrinsic toxicity and potential environmental impact. With an objective of obtaining information on the composition of printed circuit boards (PCBs) through non-invasive analysis to aid in recycling and recovery of precious waste, the goal of this paper is to propose a scheme towards the fusion of RGB and hyperspectral data in object detection. State-of-art detectors come with their own set of challenges which make them inapplicable to PCB recycling. We introduce a method which promises to achieve object detection based on multi-sensor data by utilizing the hyperspectral data to localize components and compare the results to a conventional single-sensor (RGB) based approach.

Anatomical site variation of water content in human skin measured by the Epsilon: A pilot study.

BackgroundThis pilot study aimed to investigate the anatomical site variation of water content of the stratum corneum (SC) on the body by measuring skin capacitance with the Epsilon, a new generation corneometer with multiple sensors. Secondly, values of the Epsilon were compared to values measured by conventional single sensor corneometers.MethodsThe hydration status of SC was measured in 15 healthy Caucasian volunteers with the Epsilon at five body sites (cheek, lower forearm, mid‐calf, lower back and abdomen). Transepidermal water loss (TEWL) was measured with the Aquaflux to get more insight into the condition of the skin barrier. A literature search was performed to compare Epsilon values with conventional corneometers.ResultsThe tested anatomical locations showed significant differences in water content (P < 0.001) with large interindividual variations; highest values were found in the cheek (11.64ε) and lowest values in the mid‐calf (4.43ε). No correlation between water content and TEWL was found. In general, Epsilon values were lower compared to values of conventional corneometers, with a similar trend.ConclusionThis pilot study showed significant variations in water content at different skin locations measured by the Epsilon. Moreover, the Epsilon measured consistent lower values compared to single sensor corneometers. Further validation of the device is recommended.

State Estimation based Target Tracking and Applications of Multi Sensor Data Fusion

Detection, discrimination and tracking of a target under a given dynamic environment is one of the main challenges of an integrated sensory system. A combination of two or more sensors will always provide a better position estimate rather than a single sensor. The advantages of multi-sensor data fusion over a conventional single sensor tracking are presented in this paper. Using state estimators from simple linear rustic filters to a complex nonlinear filter, the tracking of target performing three different motions with sensor noises are presented in this paper. RADAR and the infrared search and track (IRST) are the two sensors considered based on which a complete mathematical modelling and simulation of the sensor measurements and tracking methodologies are utilised. The extension of the paper also presents the image fusion techniques using a widely known technique known as principle component analysis method. The RMS errors in position as well as image error measurements are performed that shows the superiority of the multi sensor data fusion process.

Multisensor inverse synthetic aperture radar imaging and phase adjustment based on combination of sparsity and total variation

Compared with conventional single-sensor inverse synthetic aperture radar (ISAR), multisensor ISAR technique can reduce the integration time and improve the resolution of the image effectively. One solution is investigated for multisensor ISAR imaging and phase adjustment with sparse measurements via the compressed sensing (CS) framework. In most previous research of CS-based radar imaging, the target is modeled as a few strong scatterers that randomly distributed in the imaging plane and only the image domain sparsity assumption is used in image reconstruction, which will result in the degradation of image quality, especially when the measurements are limited. In practical ISAR imaging, some strong scatterers facing the radar always form flat regions in high-resolution radar imaging. Therefore, the dependence and redundancy of these scatterers can also be exploited in target reconstruction. We utilize this information and propose a multiplatform ISAR imaging method that combines the image domain sparsity and edge-preserving total variation to improve the image quality with sparse measurements. Meanwhile, an iterative minimization approach is also used to process the phase adjustment with the discontinuous echo data. Experimental results show the effectiveness of the proposed method.

Sleep monitoring sensor using flexible metal strain gauge

This paper presents a sleep monitoring sensor based on a flexible metal strain gauge. As quality of life has improved, interest in sleep quality, and related products, has increased. In this study, unlike a conventional single sensor based on a piezoelectric material, a metal strain gauge-based array sensor based on polyimide and nickel chromium (NiCr) is applied to provide movement direction, respiration, and heartbeat data as well as contact-free use by the user during sleeping. Thin-film-type resistive strain gage sensors are fabricated through the conventional flexible printed circuit board (FPCB) process, which is very useful for commercialization. The measurement of movement direction and respiratory rate during sleep were evaluated, and the heart rate data were compared with concurrent electrocardiogram (ECG) data. An algorithm for analyzing sleep data was developed using MATLAB, and the error rate was 4.2% when compared with ECG for heart rate.

Combining spatial and spectral information to estimate chlorophyll contents of crop leaves with a field imaging spectroscopy system

Imaging spectroscopy is widely used in weed recognition, pest monitoring, agricultural product quality control and other precision agricultural fields. In the present study, an in-house-designed/developed field imaging spectroscopy system (FISS, 380–870 nm) was used to obtain the imaging spectra of soybean leaves at 344 wavelengths. The spatial and spectral information including the entropy, mean reflectivity and standard deviation of the leaf images at different wavelengths were extracted; the chlorophyll content was retrieved using multiple linear regression (MLR) together with the spatial information and spectral information, and the results were compared with the results derived with the Analytical Spectral Devices (ASD, FieldSpecFR spectrometer, Analytical Spectral Devices Inc., USA) data that were generated using conventional single sensor spectrometers. The results demonstrated that the entropy, standard deviation and other features of the image were very good indicators of the leaf chlorophyll content, confirming the idea that spatial information can be used to retrieve chlorophyll content, with an accuracy equivalent to that of spectral information, and can provide information that spectral reflectivity cannot provide. Thus, integrating spatial information and spectral information can greatly improve the chlorophyll content retrieval accuracy and reduce the estimation errors by 20 %. Due to the unique measurement method and image-spectrum-in-one feature, the field imaging spectroscopy system (FISS) data can be conveniently used to achieve accurate chlorophyll content retrieval, and the retrieval error was reduced by 30–45 % compared with that for the ASD data. FISS data and the proposed method of integrating both spectral and spatial information of imaging spectroscopy have potential advantages in quantitative spectral analysis applied in agricultural biochemistry related fields.

Advanced induction motor drive control with single current sensor

This paper proposes induction motor drive control method which uses minimal number of sensors, providing only DC-link current as a feedback signal. Improved DC-link current sampling scheme and modified asymmetrical switching pattern cancels characteristic waveform errors which exist in all three reconstructed motor line-currents. Motor linecurrent harmonic content is reduced to an acceptable level, eliminating torque and speed oscillations which were inherent for conventional single sensor drives. Consequently, use of single current sensor and line-current reconstruction technique is no longer acceptable only for low and medium performance drives, but also for drives where priority is obtaining a highly accurate, stable and fast response. Proposed control algorithm is validated using induction motor drive hardware prototype based on TMS320F2812 digital signal processor.

Analysis of Dual-Sensor Stereo Geometry and Its Positioning Accuracy

This research investigated the positioning accuracy and imaging geometry of dual-sensor stereo pairs in comparison with those of conventional single-sensor stereo pairs. First, we discuss the difference between single- and dual-sensor stereo geometry and suggest that bisector elevation (BIE) and asymmetry angle, in addition to convergence angle, are important considerations for interpreting dual-sensor stereo geometry. We point out that for single-sensor BIE angles and asymmetry angles are close to ideal values, and their importance has not been highlighted. We show that dual-sensor stereo may produce small BIE angles and large asymmetry angles, and such angles may lead to weak geometry and accuracy degradation of dual-sensor stereo. Second, we compare rigorous sensor models (RSMs) with rational function models (RFMs) for handling dual-sensor stereo. For stable dual-sensor stereo, both models show similar performance. However, for dual-sensor stereo with weak geometry, RSMs produce better positioning accuracies than RFMs.

“Statistics 102” for Multisource-Multitarget Detection and Tracking

This tutorial paper summarizes the motivations, concepts and techniques of finite-set statistics (FISST), a system-level, “top-down,” direct generalization of ordinary single-sensor, single-target engineering statistics to the realm of multisensor, multitarget detection and tracking. Finite-set statistics provides powerful new conceptual and computational methods for dealing with multisensor-multitarget detection and tracking problems. The paper describes how “multitarget integro-differential calculus” is used to extend conventional single-sensor, single-target formal Bayesian motion and measurement modeling to general tracking problems. Given such models, the paper describes the Bayes-optimal approach to multisensor-multitarget detection and tracking: the multisensor-multitarget recursive Bayes filter. Finally, it describes how multitarget calculus is used to derive principled statistical approximations of this optimal filter, such as PHD filters, CPHD filters, and multi-Bernoulli filters.

"Statistics 101" for multisensor, multitarget data fusion

This tutorial summarizes the motivations, concepts, techniques, and applications of finite-set statistics (FISST), a system-level, "top-down" direct generalization of ordinary single-sensor, single-target engineering statistics to the multisensor, multitarget realm. FISST provides powerful new conceptual and computational methods for dealing with multisensor, multitarget, and multi-evidence data fusion problems. The paper begins with a broad-brush overview of the basic concepts of FISST. It describes how conventional single-sensor, single-target formal Bayesian modeling is carefully extended to general data fusion problems. We examine a simple example: joint detection and tracking of a possibly non-existent maneuvering target in heavy clutter. The tutorial concludes with a commentary on certain criticisms of FISST.

Experimental and theoretical studies on a resolution enhanced sensing system based on multisensor successive approximation

A new system for highly-accurate sensing is described that uses a multiple of inaccurate sensors. The system is based on principles of successive approximation and statistical estimation. The concept has been implemented in a microcontroller based real-time system. Experimental results for temperature measurement have demonstrated significantly improved accuracy over a conventional single-sensor system. Theoretical analysis supports experimental results. The concept is general and extends to other measurands.

A human-knowledge-based sensor implemented in an intelligent fermentation-sensor system

In many complex, ill-structured chemical and biological processes the quality variables related to the performance of a product of defined quality are hard to measure on-line. It is a fact that many processes depend on the human operator's experience and knowledge to manage unexpected situations that may occur. This paper discusses the problem of integrating human perceptual information into an objective sensor system. An intelligent fermentation-sensor system is presented, implementing the human-knowledge-based sensor in a complex operator-influenced and intuitive application. A conventional single sensor is integrated with extended computer-based time-series analysis. By extracting three statistical moments which are fused with the opinion of a human operator, we obtain an intelligent sensor system. The sensor system is applied to a bread-making fermentation process with promising results. The system is able to analyse a fermentation process and support the operator to perform the optimal decision. The sensor system can thus be used to facilitate on-line supervisory control, especially to detect and predict unexpected property variations in the measured process.