ALBERT

Beschreibung: In machine learning and data science, feature selection is considered as a crucial step of data preprocessing. When we directly apply the raw data for classification or clustering purposes, sometimes we observe that the learning algorithms do not perform well. One possible reason for this is the presence of redundant, noisy, and non-informative features or attributes in the datasets. Hence, feature selection methods are used to identify the subset of relevant features that can maximize the model performance. Moreover, due to reduction in feature dimension, both training time and storage required by the model can be reduced as well. In this paper, we present a tri-stage wrapper-filter-based feature selection framework for the purpose of medical report-based disease detection. In the first stage, an ensemble was formed by four filter methods—Mutual Information, ReliefF, Chi Square, and Xvariance—and then each feature from the union set was assessed by three classification algorithms—support vector machine, naïve Bayes, and k-nearest neighbors—and an average accuracy was calculated. The features with higher accuracy were selected to obtain a preliminary subset of optimal features. In the second stage, Pearson correlation was used to discard highly correlated features. In these two stages, XGBoost classification algorithm was applied to obtain the most contributing features that, in turn, provide the best optimal subset. Then, in the final stage, we fed the obtained feature subset to a meta-heuristic algorithm, called whale optimization algorithm, in order to further reduce the feature set and to achieve higher accuracy. We evaluated the proposed feature selection framework on four publicly available disease datasets taken from the UCI machine learning repository, namely, arrhythmia, leukemia, DLBCL, and prostate cancer. Our obtained results confirm that the proposed method can perform better than many state-of-the-art methods and can detect important features as well. Less features ensure less medical tests for correct diagnosis, thus saving both time and cost.

Beschreibung: With the rapid development of deep learning, computer vision has assisted in solving a variety of problems in engineering construction. However, very few computer vision-based approaches have been proposed on work productivity’s evaluation. Therefore, taking a super high-rise project as a research case, using the detected object information obtained by a deep learning algorithm, a computer vision-based method for evaluating the productivity of assembling reinforcement is proposed. Firstly, a detector that can accurately distinguish various entities related to assembling reinforcement based on CenterNet is established. DLA34 is selected as the backbone. The mAP reaches 0.9682, and the speed of detecting a single image can be as low as 0.076 s. Secondly, the trained detector is used to detect the video frames, and images with detected boxes and documents with coordinates can be obtained. The position relationship between the detected work objects and detected workers is used to determine how many workers (N) have participated in the task. The time (T) to perform the process can be obtained from the change of coordinates of the work object. Finally, the productivity is evaluated according to N and T. The authors use four actual construction videos for validation, and the results show that the productivity evaluation is generally consistent with the actual conditions. The contribution of this research to construction management is twofold: On the one hand, without affecting the normal behavior of workers, a connection between construction individuals and work object is established, and the work productivity evaluation is realized. On the other hand, the proposed method has a positive effect on improving the efficiency of construction management.

Beschreibung: Physical inactivity increases the risk of many adverse health conditions, including the world’s major non-communicable diseases, such as coronary heart disease, type 2 diabetes, and breast and colon cancers, shortening life expectancy. There are minimal medical care and personal trainers’ methods to monitor a patient’s actual physical activity types. To improve activity monitoring, we propose an artificial-intelligence-based approach to classify physical movement activity patterns. In more detail, we employ two deep learning (DL) methods, namely a deep feed-forward neural network (DNN) and a deep recurrent neural network (RNN) for this purpose. We evaluate the two models on two physical movement datasets collected from several volunteers who carried tri-axial accelerometer sensors. The first dataset is from the UCI machine learning repository, which contains 14 different activities-of-daily-life (ADL) and is collected from 16 volunteers who carried a single wrist-worn tri-axial accelerometer. The second dataset includes ten other ADLs and is gathered from eight volunteers who placed the sensors on their hips. Our experiment results show that the RNN model provides accurate performance compared to the state-of-the-art methods in classifying the fundamental movement patterns with an overall accuracy of 84.89% and an overall F1-score of 82.56%. The results indicate that our method provides the medical doctors and trainers a promising way to track and understand a patient’s physical activities precisely for better treatment.

Beschreibung: The purpose of this study was to investigate the feasibility of a time-of-flight (TOF) brain positron emission tomography (PET) providing high-quality images. It consisted of 30 detector blocks arranged in a ring with a diameter of 257 mm and an axial field of view of 52.2 mm. Each detector block was composed of two detector modules and two application-specific integrated circuit (ASIC) chips. The detector module was composed of an 8 × 8 array of 3 × 3 mm2 multi-pixel photon counters and an 8 × 8 array of 3.11 × 3.11 × 15 mm3 lutetium yttrium oxyorthosilicate scintillators. The 64-channel individual readout ASIC was used to acquire the position, energy, and time information of a detected gamma ray. A coincidence timing resolution of 187 ps full width at half maximum (FWHM) was achieved using a pair of channels of two detector modules. The energy resolution and spatial resolution were 6.6 ± 0.6% FWHM (without energy nonlinearity correction) and 2.5 mm FWHM, respectively. The results of this study demonstrate that the developed TOF brain PET could provide excellent performance, allowing for a reduction in radiation dose or scanning time for brain imaging due to improved sensitivity and signal-to-noise ratio.

Beschreibung: Detecting trace amounts of explosives to ensure personal safety is important, and this is possible by using laser-based spectroscopy techniques. We performed surface-enhanced Raman scattering (SERS) using plasmonic nanogap substrates for the solution phase detection of some nitro-based compounds, taking advantage of the hot spot at the nanogap. An excitation wavelength of 785 nm with an incident power of as low as ≈0.1 mW was used to excite the nanogap substrates. Since both RDX and PETN cannot be dissolved in water, acetone was used as a solvent. TNT was dissolved in water as well as in hexane. The main SERS peaks of TNT, RDX, and PETN were clearly observed down to the order of picomolar concentration. The variations in SERS spectra observed from different explosives can be useful in distinguishing and identifying different nitro-based compounds. This result indicates that our nanogap substrates offer an effective approach for explosives identification.

Beschreibung: Rayleigh waves are very useful for ultrasonic nondestructive evaluation of structural and mechanical components. Nonlinear Rayleigh waves have unique sensitivity to the early stages of material degradation because material nonlinearity causes distortion of the waveforms. The self-interaction of a sinusoidal waveform causes second harmonic generation, while the mutual interaction of waves creates disturbances at the sum and difference frequencies that can potentially be detected with minimal interaction with the nonlinearities in the sensing system. While the effect of surface roughness on attenuation and dispersion is well documented, its effects on the nonlinear aspects of Rayleigh wave propagation have not been investigated. Therefore, Rayleigh waves are sent along aluminum surfaces having small, but different, surface roughness values. The relative nonlinearity parameter increased significantly with surface roughness (average asperity heights 0.027–3.992 μm and Rayleigh wavelengths 0.29–1.9 mm). The relative nonlinearity parameter should be decreased by the presence of attenuation, but here it actually increased with roughness (which increases the attenuation). Thus, an attenuation-based correction was unsuccessful. Since the distortion from material nonlinearity and surface roughness occur over the same surface, it is necessary to make material nonlinearity measurements over surfaces having the same roughness or in the future develop a quantitative understanding of the roughness effect on wave distortion.

Beschreibung: Deep learning technology has improved the performance of vision-based action recognition algorithms, but such methods require a large number of labeled training datasets, resulting in weak universality. To address this issue, this paper proposes a novel self-deployable ubiquitous action recognition framework that enables a self-motivated user to bootstrap and deploy action recognition services, called FOLLOWER. Our main idea is to build a “fingerprint” library of actions based on a small number of user-defined sample action data. Then, we use the matching method to complete action recognition. The key step is how to construct a suitable “fingerprint”. Thus, a pose action normalized feature extraction method based on a three-dimensional pose sequence is designed. FOLLOWER is mainly composed of the guide process and follow the process. Guide process extracts pose action normalized feature and selects the inner class central feature to build a “fingerprint” library of actions. Follow process extracts the pose action normalized feature in the target video and uses the motion detection, action filtering, and adaptive weight offset template to identify the action in the video sequence. Finally, we collect an action video dataset with human pose annotation to research self-deployable action recognition and action recognition based on pose estimation. After experimenting on this dataset, the results show that FOLLOWER can effectively recognize the actions in the video sequence with recognition accuracy reaching 96.74%.

Beschreibung: The control of glucose concentration is a crucial factor in clinical diagnosis and the food industry. Electrochemical biosensors based on reduced graphene oxide (rGO) and conducting polymers have a high potential for practical application. A novel thermal reduction protocol of graphene oxide (GO) in the presence of malonic acid was applied for the synthesis of rGO. The rGO was characterized by scanning electron microscopy, X-ray diffraction analysis, Fourier-transform infrared spectroscopy, and Raman spectroscopy. rGO in combination with polyaniline (PANI), Nafion, and glucose oxidase (GOx) was used to develop an amperometric glucose biosensor. A graphite rod (GR) electrode premodified with a dispersion of PANI nanostructures and rGO, Nafion, and GOx was proposed as the working electrode of the biosensor. The optimal ratio of PANI and rGO in the dispersion used as a matrix for GOx immobilization was equal to 1:10. The developed glucose biosensor was characterized by a wide linear range (from 0.5 to 50 mM), low limit of detection (0.089 mM), good selectivity, reproducibility, and stability. Therefore, the developed biosensor is suitable for glucose determination in human serum. The PANI nanostructure and rGO dispersion is a promising material for the construction of electrochemical glucose biosensors.

Beschreibung: A new optical fiber sensor based on the fluorescence lifetime was prepared for specific detection of sulfate ion concentration, where 1,1′-(anthracene-9,10-diylbis(methylene))bis(3-(dodecylcarbamoyl)pyridin-1-ium) acted as the sulfate fluorescent probe. The probe was immobilized in a porous cellulose acetate membrane to form the sensitive membrane by the immersion precipitation method, and polyethylene glycol 400 acted as a porogen. The sensing principle was proven, as a sulfate ion could form a complex with the probe through a hydrogen bond, which led to structural changes and fluorescence for the probe. The signals of the fluorescence lifetime data were collected by the lock-in amplifier and converted into the phase delay to realize the detection of sulfate ions. Based on the phase-modulated fluorometry, the relationship between the phase delay of the probe and the sulfate ion concentration was described in the range from 2 to 10 mM. The specificity and response time of this optical fiber sensor were also researched.

Beschreibung: The actively heated fiber optics (AHFO) technique has the potential to measure soil water at high spatial and temporal resolutions, and thus it can bridge the measurement gap from point to large scales. However, the availability of power might restrict its use, since high power is required to heat long fiber optic cables under field conditions; this can be a challenge for long-term soil water monitoring under field conditions. This study investigated the performance of different heating strategies (power intensity and heating duration) on soil water measurement by the AHFO technique on three different textured soils. Different heating strategies: high power–short pulses (20 Wm−1–3 min), low power–short pulses (10 Wm−1–3 min, 5 Wm−1–3 min, 2.5 Wm−1–3 min) and low power–long pulses (10 Wm−1–5 min, 5 Wm−1–10 min, 2.5 Wm−1–15 min) were tested using laboratory soil columns. The study compared the sensitivity of the thermal response, NTcum to volumetric water content (VWC) and the predictive error of different heating strategies and soils. Results of this study showed that the sensitivity of NTcum increased and the predictive error decreased with increasing power intensity, irrespective of the soil type. Low power–short heat pulses such as 5 Wm−1–3 min and 2.5 Wm−1–3 min produced high predictive errors, RMSE of 5–6% and 6–7%, respectively. However, extending the heating duration was effective in reducing the error for both 10 and 5 Wm−1 power intensities, but not for the 2.5 Wm−1. The improvement was particularly noticeable in 5 Wm−1 –10 min; it reduced the RMSE by 1.5% (sand and clay loam) and 2.73% (sandy loam). Overall, the results of this study suggested that extending the heating duration of 10 and 5 Wm−1 power intensities can improve the sensitivity of the thermal response and predictive accuracy of the estimated soil water content (SWC). The results are particularly important for field applications of the AHFO technique, which can be limited by the availability of high power, which restricts the use of 20 Wm−1. For example, 5 Wm−1–10 min improved the predictive accuracy to 3–4%, which has the potential to be used for validating soil water estimations at satellite footprint scales. However, the effects of diurnal temperature variations should also be considered, particularly when using low power intensity such as 5 Wm−1 in surface soils under field conditions.

Beschreibung: Electrocardiogram (ECG) signals play a vital role in diagnosing and monitoring patients suffering from various cardiovascular diseases (CVDs). This research aims to develop a robust algorithm that can accurately classify the electrocardiogram signal even in the presence of environmental noise. A one-dimensional convolutional neural network (CNN) with two convolutional layers, two down-sampling layers, and a fully connected layer is proposed in this work. The same 1D data was transformed into two-dimensional (2D) images to improve the model’s classification accuracy. Then, we applied the 2D CNN model consisting of input and output layers, three 2D-convolutional layers, three down-sampling layers, and a fully connected layer. The classification accuracy of 97.38% and 99.02% is achieved with the proposed 1D and 2D model when tested on the publicly available Massachusetts Institute of Technology-Beth Israel Hospital (MIT-BIH) arrhythmia database. Both proposed 1D and 2D CNN models outperformed the corresponding state-of-the-art classification algorithms for the same data, which validates the proposed models’ effectiveness.

Beschreibung: The emergence of very high-resolution (VHR) satellite imagery (less than 1 m spatial resolution) is creating new opportunities within the fields of ecology and conservation biology. The advancement of sub-meter resolution imagery has provided greater confidence in the detection and identification of features on the ground, broadening the realm of possible research questions. To date, VHR imagery studies have largely focused on terrestrial environments; however, there has been incremental progress in the last two decades for using this technology to detect cetaceans. With advances in computational power and sensor resolution, the feasibility of broad-scale VHR ocean surveys using VHR satellite imagery with automated detection and classification processes has increased. Initial attempts at automated surveys are showing promising results, but further development is necessary to ensure reliability. Here we discuss the future directions in which VHR satellite imagery might be used to address urgent questions in whale conservation. We highlight the current challenges to automated detection and to extending the use of this technology to all oceans and various whale species. To achieve basin-scale marine surveys, currently not feasible with any traditional surveying methods (including boat-based and aerial surveys), future research requires a collaborative effort between biology, computation science, and engineering to overcome the present challenges to this platform’s use.

Beschreibung: The internet-of-things (IoT) is expected to have a transformative impact in several different domains, including energy management in smart grids, manufacturing, transportation, smart cities and communities, smart food and farming, and healthcare. To this direction, the maintenance cost of IoT deployments has been identified as one of the main challenges, which is directly related to energy efficiency and autonomy of IoT solutions. In order to increase the energy sustainability of next-generation IoT, wireless power transfer (WPT) emerged as a promising technology; however, its effectiveness is hindered as the distance between the base station and the wireless powered IoT devices increases. To counter this effect, decentralized approaches based on the use of distributed densely deployed remote radio heads (RRHs) can be utilized to diminish the distance between the transmitting and the receiving nodes. A trade-off ensues from the use of RRHs as power beacons (PBs) or access points (APs) that enable either energy transfer during downlink or information reception during uplink, respectively. To balance this trade-off, in this work, the maximization of the ergodic rate in ultra-dense wireless powered networks is investigated. In more detail, three different protocols are introduced, optimized, and compared to each other: density splitting, time splitting, and hybrid time and density splitting, which are based on the optimization of the portion of the number of RRHs that are employed as PBs or APs at a specific time instance. Additionally, two different policies are taken into account regarding the PBs’ power constraint. The formulated problems that correspond to the combination of the proposed protocols with each of the two considered power constraint policies are optimally solved by using convex optimization tools and closed-form solutions are derived that result to useful insights. Finally, numerical results are provided, which illustrate the ergodic rate achieved by each of the proposed protocols and offer interesting conclusions regarding their comparison, which are directly linked to design guidelines and the required capital and operational expenses.

Beschreibung: Phacoemulsification is a widely used surgical method in cataract surgery with a high energy ultrasound source. The viscoelastic is considered to be tissue protective. The aim of this study is to investigate during surgery the impact of using viscoelastic versus no viscoelastic on clinical outcomes, potential complications and effect on endothelial cell density. The study group included 64 patients, who were subjected to phacoemulsification using balanced salt solution (BSS). Control group consisted of 62 patients, who underwent phacoemulsification using Hyaloronic acid 1% Healon 1%. Student’s t-test was applied for statistical analysis. The simulations of temperature changes during phacoemulsification were performed by COMSOL Multiphysics software. In the BSS group, a mean endothelial cell loss (ECL) of 4.5% was measured one month postoperatively, while in the Healon group ECL was 5.3%. Data analysis showed no significant difference in ECL between the groups (Student’s t-test, p = 0.8). No significant difference was observed in endothelial cell morphology and IOP between the two groups pre- and postoperatively (all p 〉 0.05). The modeling of thermo fluid dynamics showed that the heating of the cornea is slightly less when Healon was used as irrigation fluid. The phacoemulsification technique can be performed by an experienced surgeon with viscoelastics or continuous anterior chamber (AC) irrigation on the same level of safety regarding endothelial cell damage, providing equally satisfying clinical outcomes.

Beschreibung: In this paper, a joint multiple-input multiple-output (MIMO OFDM) radar and communication (RadCom) system is proposed, in which orthogonal frequency division multiplexing (OFDM) waveforms carrying data to be transmitted to the information receiver are exploited to get high-resolution radar images at the RadCom platform. Specifically, to get two-dimensional (i.e., range and azimuth angle) radar images with high resolution, a compressive sensing-based imaging algorithm is proposed that is applicable to the signal received through multiple receive antennas. Because both the radar imaging performance (i.e., the mean square error of the radar image) and the communication performance (i.e., the achievable rate) are affected by the subcarrier allocation across multiple transmit antennas, by analyzing both radar imaging and communication performances, we also propose a subcarrier allocation strategy such that a high achievable rate is obtained without sacrificing the radar imaging performance.

Beschreibung: This research is focused on searching for frequency and noise characteristics for available GNSS (Global Navigation Satellite Systems). The authors illustrated frequency stability and noise characteristics for a selected set of data from four different GNSS systems. For this purpose, 30-s-interval clock corrections were used for the GPS weeks 1982–2034 (the entirety of 2018). Firstly, phase data (raw clock corrections) were preprocessed for shifts and removal of outliers; GLONASS and GPS satellites characterize a smaller number of outliers than BeiDou and Galileo clock products. Secondly, frequency and Hadamard deviation were calculated. This study concludes that the stability of GPS and Galileo is better than that of BDS (BeiDou Navigation Satellite System) and GLONASS. Regarding noise, the GPS, Galileo, and BDS clocks are affected by the random walk modulation noise (RWFM), flashing frequency modulation noise (FFM), and white frequency modulation noise (WFM), whereas the GLONASS clocks are mainly affected only by WFM.

Beschreibung: In this paper, we propose a new approach to the attitude control of quadrotors, by which angular velocity measurements or a model-based observer reconstructing the angular velocity are not needed. The proposed approach is based on recent stability results obtained for nonlinear negative imaginary systems. In specific, through an inner-outer loop method, we establish the nonlinear negative imaginary property of the quadrotor rotational subsystem. Then, a strictly negative imaginary controller is synthesized using the nonlinear negative imaginary results. This guarantees the robust asymptotic stability of the attitude of the quadrotor in the face of modeling uncertainties and external disturbances. First simulation results underline the effectiveness of the proposed attitude control approach are presented.

Beschreibung: Weather change such as raining is a crucial factor to cause traffic congestion, especially in metropolises with the limited sewer system infrastructures. Identifying the roads which are sensitive to weather changes, defined as weather-sensitive roads (WSR), can facilitate the infrastructure development. In the literature, little research focused on studying weather factors of developing countries that might have deficient infrastructures. In this research, to fill the gap, the real-world data associating with Jakarta, Indonesia, was studied to identify WSR based on smartphone sensor data, real-time weather information, and road characteristics datasets. A spatial-temporal congestion speed matrix (STC) was proposed to illustrate traffic speed changes over time. Under the proposed STC, a sequential clustering and classification framework was applied to identify the WSR in terms of traffic speed. In this work, the causes of WSR were evaluated based on the variables’ importance of the classification method. The experimental results show that the proposed method can cluster the roads according to the pattern changes in the traffic speed caused by weather change. Based on the results, we found that the distances to shopping malls, mosques, schools, and the roads’ altitude, length, width, and the number of lanes are highly correlated to WSR in Jakarta.

Beschreibung: High stress levels and sleep deprivation may cause several mental or physical health issues, such as depression, impaired memory, decreased motivation, obesity, etc. The COVID-19 pandemic has produced unprecedented changes in our lives, generating significant stress, and worries about health, social isolation, employment, and finances. To this end, nowadays more than ever, it is crucial to deliver solutions that can help people to manage and control their stress, as well as to reduce sleep disturbances, so as to improve their health and overall quality of life. Technology, and in particular Ambient Intelligence Environments, can help towards that direction, when considering that they are able to understand the needs of their users, identify their behavior, learn their preferences, and act and react in their interest. This work presents two systems that have been designed and developed in the context of an Intelligent Home, namely CaLmi and HypnOS, which aim to assist users that struggle with stress and poor sleep quality, respectively. Both of the systems rely on real-time data collected by wearable devices, as well as contextual information retrieved from the ambient facilities of the Intelligent Home, so as to offer appropriate pervasive relaxation programs (CaLmi) or provide personalized insights regarding sleep hygiene (HypnOS) to the residents. This article will describe the design process that was followed, the functionality of both systems, the results of the user studies that were conducted for the evaluation of their end-user applications, and a discussion about future plans.

Beschreibung: Recent studies have applied the superior performance of deep learning to mobile devices, and these studies have enabled the running of the deep learning model on a mobile device with limited computing power. However, there is performance degradation of the deep learning model when it is deployed in mobile devices, due to the different sensors of each device. To solve this issue, it is necessary to train a network model specific to each mobile device. Therefore, herein, we propose an acceleration method for on-device learning to mitigate the device heterogeneity. The proposed method efficiently utilizes unified memory for reducing the latency of data transfer during network model training. In addition, we propose the layer-wise processor selection method to consider the latency generated by the difference in the processor performing the forward propagation step and the backpropagation step in the same layer. The experiments were performed on an ODROID-XU4 with the ResNet-18 model, and the experimental results indicate that the proposed method reduces the latency by at most 28.4% compared to the central processing unit (CPU) and at most 21.8% compared to the graphics processing unit (GPU). Through experiments using various batch sizes to measure the average power consumption, we confirmed that device heterogeneity is alleviated by performing on-device learning using the proposed method.

Beschreibung: Acoustic emission (AE) monitoring has become an optional technology to quantify slope deformation. However, there are still challenges in developing generic AE interpretation strategies. Dynamics and kinematics models are two physical methods for analysing slope stability, which appear to improve the interpretability of AE monitoring data. The aim of this study is to explore the change patterns and interrelations of dynamics, kinematics, and AE measurements using a model test and physical analysis, to further understand the development process of a progressive landslide. A model test is designed based on the kinematics model of landslide three-stage deformation. An equation between factor of safety (FoS) and thrust is proposed based on the mechanical model of a landslide test. There is a clear correspondence between the displacement and inverse velocity during the deformation-controlled process. Relationships are uncovered between the thrust and FoS as well as the thrust and acceleration. As a characteristic parameter of AE, ring down count (RDC) is able to quantify the deformation process of the soil slope. Moreover, acceleration and RDC can reflect the sudden change of the slope state and, hence, can be effective indicators for the early warning in a progressive landslide.

Beschreibung: Fatigued driving is one of the main causes of traffic accidents. The electroencephalogram (EEG)-based mental state analysis method is an effective and objective way of detecting fatigue. However, as EEG shows significant differences across subjects, effectively “transfering” the EEG analysis model of the existing subjects to the EEG signals of other subjects is still a challenge. Domain-Adversarial Neural Network (DANN) has excellent performance in transfer learning, especially in the fields of document analysis and image recognition, but has not been applied directly in EEG-based cross-subject fatigue detection. In this paper, we present a DANN-based model, Generative-DANN (GDANN), which combines Generative Adversarial Networks (GAN) to enhance the ability by addressing the issue of different distribution of EEG across subjects. The comparative results show that in the analysis of cross-subject tasks, GDANN has a higher average accuracy of 91.63% in fatigue detection across subjects than those of traditional classification models, which is expected to have much broader application prospects in practical brain–computer interaction (BCI).

Beschreibung: Object detection is an indispensable part of autonomous driving. It is the basis of other high-level applications. For example, autonomous vehicles need to use the object detection results to navigate and avoid obstacles. In this paper, we propose a multi-scale MobileNeck module and an algorithm to improve the performance of an object detection model by outputting a series of Gaussian parameters. These Gaussian parameters can be used to predict both the locations of detected objects and the localization confidences. Based on the above two methods, a new confidence-aware Mobile Detection (MobileDet) model is proposed. The MobileNeck module and loss function are easy to conduct and integrate with Generalized-IoU (GIoU) metrics with slight changes in the code. We test the proposed model on the KITTI and VOC datasets. The mean Average Precision (mAP) is improved by 3.8 on the KITTI dataset and 2.9 on the VOC dataset with less resource consumption.

Beschreibung: The preservation of cultural heritage assets of all kind is an important task for modern civilizations. This also includes tools and instruments that have been used in the previous decades and centuries. Along with the industrial revolution 200 years ago, mechanical and electrical technologies emerged, together with optical instruments. In the meantime, it is not only museums who showcase these developments, but also companies, universities, and private institutions. Gyroscopes are fascinating instruments with a history dating back 200 years. When J.G.F. Bohnenberger presented his machine to his students in 1810 at the University of Tuebingen, Germany, nobody could have foreseen that this fascinating development would be used for complex orientation and positioning. At the University of Stuttgart, Germany, a collection of 160 exhibits is available and in transition towards their sustainable future. Here, the systems are digitized in 2D, 2.5D, and 3D and are made available for a worldwide community using open access platforms. The technologies being used are computed tomography, computer vision, endoscopy, and photogrammetry. We present a novel workflow for combining voxel representations and colored point clouds, to create digital twins of the physical objects with 0.1 mm precision. This has not yet been investigated and is therefore pioneering work. Advantages and disadvantages are discussed and suggested work for the near future is outlined in this new and challenging field of tech heritage digitization.

Beschreibung: With the development of smart vehicles and various vehicular applications, Vehicular Edge Computing (VEC) paradigm has attracted from academic and industry. Compared with the cloud computing platform, VEC has several new features, such as the higher network bandwidth and the lower transmission delay. Recently, vehicular computation-intensive task offloading has become a new research field for the vehicular edge computing networks. However, dynamic network topology and the bursty computation tasks offloading, which causes to the computation load unbalancing for the VEC networking. To solve this issue, this paper proposed an optimal control-based computing task scheduling algorithm. Then, we introduce software defined networking/OpenFlow framework to build a software-defined vehicular edge networking structure. The proposed algorithm can obtain global optimum results and achieve the load-balancing by the virtue of the global load status information. Besides, the proposed algorithm has strong adaptiveness in dynamic network environments by automatic parameter tuning. Experimental results show that the proposed algorithm can effectively improve the utilization of computation resources and meet the requirements of computation and transmission delay for various vehicular tasks.

Beschreibung: Sleep disruption from causes, such as changes in lifestyle, stress from aging, family issues, or life pressures are a growing phenomenon that can lead to serious health problems. As such, sleep disorders need to be identified and addressed early on. In recent years, studies have investigated sleep patterns through body movement information collected by wristwatch-type devices or cameras. However, these methods capture only the individual’s awake and sleep states and lack sufficient information to identify specific sleep stages. The aim of this study was to use a 3-axis accelerometer attached to an individual’s head to capture information that can identify three specific sleep stages: rapid eye movement (REM) sleep, light sleep, and deep sleep. These stages are measured by heart rate features captured by a ballistocardiogram and body movement. The sleep experiment was conducted for two nights among eight healthy adult men. According to the leave-one-out cross-validation results, the F-scores were: awake 76.6%, REM sleep 52.7%, light sleep 78.2%, and deep sleep 67.8%. The accuracy was 74.6% for the four estimates. This proposed measurement system was able to estimate the sleep stages with high accuracy simply by using the acceleration in the individual’s head.

Beschreibung: When a wireless sensor node’s wireless communication fails after being deployed in an inaccessible area, the lost node cannot be repaired through a debugging interaction that relies on that communication. Visible light communication (VLC) as a supplement of radio wave communication can improve the transmission security at the physical layer due to its unidirectional propagation characteristic. Therefore, we implemented a VLC-based hybrid communication debugging system (HCDS) based on VLC using smartphone and sensor node. For the system’s downlink, the smartphone is taken as the VLC gateway and sends the debugging codes to the sensor node by the flashlight. To improve the transmission efficiency of the downlink, we also propose a new coding method for source coding and channel coding, respectively. For the source coding, we analyze the binary instructions and compress the operands using bitmask techniques. The average compression rate of the binary structure reaches 84.11%. For the channel coding, we optimize dual-header pulse interval (DH-PIM) and propose overlapped DH-PIM (ODH-PIM) by introducing a flashlight half-on state. The flashlight half-on state can improve the representation capability of individual symbols. For the uplink of HCDS, we use the onboard LED of the sensor node to transmit feedback debugging information to the smartphone. At the same time, we design a novel encoding format of DH-PIM to optimize uplink transmission. Experimental results show that the optimized uplink transmission time and BER are reduced by 10.71% and 22%, compared with the original DH-PIM.

Beschreibung: Fringe projection profilometry in combination with other optical measuring technologies has established itself over the last decades as an essential complement to conventional, tactile measuring devices. The non-contact, holistic reconstruction of complex geometries within fractions of a second in conjunction with the lightweight and transportable sensor design open up many fields of application in production metrology. Furthermore, triangulation-based measuring principles feature good scalability, which has led to 3D scanners for various scale ranges. Innovative and modern production processes, such as sheet-bulk metal forming, thus, utilize fringe projection profilometry in many respects to monitor the process, quantify possible wear and improve production technology. Therefore, it is essential to identify the appropriate 3D scanner for each application and to properly evaluate the acquired data. Through precise knowledge of the measurement volume and the relative uncertainty with respect to the specimen and scanner position, adapted measurement strategies and integrated production concepts can be realized. Although there are extensive industrial standards and guidelines for the quantification of sensor performance, evaluation and tolerancing is mainly global and can, therefore, neither provide assistance in the correct, application-specific positioning and alignment of the sensor nor reflect the local characteristics within the measuring volume. Therefore, this article compares fringe projection systems across various scale ranges by positioning and scanning a calibrated sphere in a high resolution grid.

Beschreibung: Indoor localization is an enabling technology for pervasive and mobile computing applications. Although different technologies have been proposed for indoor localization, Wi-Fi fingerprinting is one of the most used techniques due to the pervasiveness of Wi-Fi technology. Most Wi-Fi fingerprinting localization methods presented in the literature are discriminative methods. We present a generative method for indoor localization based on Wi-Fi fingerprinting. The Received Signal Strength Indicator received from a Wireless Access Point is modeled by a hidden Markov model. Unlike other algorithms, the use of a hidden Markov model allows ours to take advantage of the temporal autocorrelation present in the Wi-Fi signal. The algorithm estimates the user’s location based on the hidden Markov model, which models the signal and the forward algorithm to determine the likelihood of a given time series of Received Signal Strength Indicators. The proposed method was compared with four other well-known Machine Learning algorithms through extensive experimentation with data collected in real scenarios. The proposed method obtained competitive results in most scenarios tested and was the best method in 17 of 60 experiments performed.

Beschreibung: Millimeter-wave (W-band) radar measurements were taken for two maritime targets instrumented with attitude and heading reference systems (AHRSs) in a littoral environment with the aim of developing a multiaspect classifier. The focus was on resource-limited implementations such as short-range, tactical, unmanned aircraft systems (UASs) and dealing with limited and imbalanced datasets. Radar imaging and preprocessing consisted of recording high-resolution range profiles (HRRPs) and performing range alignment using peak detection and fast Fourier transforms (FFTs). HRRPs were used because of their simplicity, reliability, and speed. The features used were fixed-length, frequency domain range profiles. Two linear support vector machine (SVM)-based classifiers were developed which both yielded excellent results in their general forms and were simple to implement. The first approach utilized the positive predictive value (PPV) and negative predictive value (NPV) statistics of the SVM directly to generate target probabilities and consequently determine the optimal aspect transitions for classification. The second approach used the Kolmogorov–Smirnov test for dimensionality reduction, followed by concatenating feature vectors across several aspects. The latter approach is particularly well-suited to resource-constrained scenarios, potentially allowing for retraining and updating in the field.

Beschreibung: We investigated agent-based model simulations that mimic an ant transportation system to analyze the cooperative perception and communication in the system. On a trail, ants use cooperative perception through chemotaxis to maintain a constant average velocity irrespective of their density, thereby avoiding traffic jams. Using model simulations and approximate mathematical representations, we analyzed various aspects of the communication system and their effects on cooperative perception in ant traffic. Based on the analysis, insights about the cooperative perception of ants which facilitate decentralized self-organization is presented. We also present values of communication-parameters in ant traffic, where the system conveys traffic conditions to individual ants, which ants use to self-organize and avoid traffic-jams. The mathematical analysis also verifies our findings and provides a better understanding of various model parameters leading to model improvements.

Beschreibung: Most indoor positioning systems require calibration before use. Fingerprinting requires the construction of a signal strength map, while ranging systems need the coordinates of the beacons. Calibration approaches exist for positioning systems that use Wi-Fi, radio frequency identification or ultrawideband. However, few examples are available for the calibration of visible light positioning systems. Most works focused on obtaining the channel model parameters or performed a calibration based on known receiver locations. In this paper, we describe an improved procedure that uses a mobile robot for data collection and is able to obtain a map of the environment with the beacon locations and their identities. Compared to previous work, the error is almost halved. Additionally, this approach does not require prior knowledge of the number of light sources or the receiver location. We demonstrate that the system performs well under a wide range of lighting conditions and investigate the influence of parameters such as the robot trajectory, camera resolution and field of view. Finally, we also close the loop between calibration and positioning and show that our approach has similar or better accuracy than manual calibration.

Beschreibung: This paper is concerned with relative localization-based optimal area coverage placement using multiple unmanned aerial vehicles (UAVs). It is assumed that only one of the UAVs has its global position information before performing the area coverage task and that ranging measurements can be obtained among the UAVs by using ultra-wide band (UWB) sensors. In this case, multi-UAV relative localization and cooperative coverage control have to be run simultaneously, which is a quite challenging task. In this paper, we propose a single-landmark-based relative localization algorithm, combined with a distributed coverage control law. At the same time, the optimal multi-UAV placement problem was formulated as a quadratic programming problem by compromising between optimal relative localization and optimal coverage control and was solved by using Sequential Quadratic Programming (SQP) algorithms. Simulation results show that our proposed method can guarantee that a team of UAVs can efficiently localize themselves in a cooperative manner and, at the same time, complete the area coverage task.

Beschreibung: Recently, formation flying of multiple unmanned aerial vehicles (UAVs) found numerous applications in various areas such as surveillance, industrial automation and disaster management. The accuracy and reliability for performing group tasks by multiple UAVs is highly dependent on the applied control strategy. The formation and trajectories of multiple UAVs are governed by two separate controllers, namely formation and trajectory tracking controllers respectively. In presence of environmental effects, disturbances due to wind and parametric uncertainties, the controller design process is a challenging task. This article proposes a robust adaptive formation and trajectory tacking control of multiple quad-rotor UAVs using super twisting sliding mode control method. In the proposed design, Lyapunov function-based adaptive disturbance estimators are used to compensate for the effects of external disturbances and parametric uncertainties. The stability of the proposed controllers is guaranteed using Lyapunov theorems. Two variants of the control schemes, namely fixed gain super twisting SMC (STSMC) and adaptive super twisting SMC (ASTSMC) are tested using numerical simulations performed in MATLAB/Simulink. From the results presented, it is verified that in presence of disturbances, the proposed ASTSMC controller exhibits enhanced robustness as compared to the fixed gain STSMC.