A Real-Time Complex Road AI Perception Based on 5G-V2X for Smart City Security

The Internet of Vehicles and information security are key components of a smart city. Real-time road perception is one of the most difficult tasks. Traditional detection methods require manual adjustment of parameters, which is difficult, and is susceptible to interference from object occlusion, light changes, and road wear. Designing a robust road perception algorithm is still challenging. On this basis, we combine artificial intelligence algorithms and the 5G-V2X framework to propose a real-time road perception method. First, an improved model based on Mask R-CNN is implemented to improve the accuracy of detecting lane line features. Then, the linear and polynomial fitting methods of feature points in different fields of view are combined. Finally, the optimal parameter equation of the lane line can be obtained. We tested our method in complex road scenes. Experimental results show that, combined with 5G-V2X, this method ultimately has a faster processing speed and can sense road conditions robustly under various complex actual conditions.

Testing the accuracy of 3D automatic landmarking via genome-wide association studies

Various advances in 3D automatic phenotyping and landmark-based geometric morphometric methods have been made. While it is generally accepted that automatic landmarking compromises the capture of the biological variation, no studies have directly tested the actual impact of such landmarking approaches in analyses requiring a large number of specimens and for which the precision of phenotyping is crucial to extract an actual biological signal adequately. Here, we use a recently developed 3D atlas-based automatic landmarking method to test its accuracy in detecting QTLs associated with craniofacial development of the house mouse skull and lower jaws for a large number of specimens (circa 700) that were previously phenotyped via a semi-automatic landmarking method complemented with manual adjustment. We compare both landmarking methods with univariate and multivariate mapping of the skull and the lower jaws. We find that most significant SNPs and QTLs are not recovered based on the data derived from the automatic landmarking method. Our results thus confirm the notion that information is lost in the automated landmarking procedure although somewhat dependent on the analyzed structure. The automatic method seems to capture certain types of structures slightly better, such as lower jaws whose shape is almost entirely summarized by its outline and could be assimilated as a 2D flat object. By contrast, the more apparent 3D features exhibited by a structure such as the skull are not adequately captured by the automatic method. We conclude that using 3D atlas-based automatic landmarking methods requires careful consideration of the experimental question.

Multi-Incidence Holographic Profilometry for Large Gradient Surfaces with Sub-Micron Focusing Accuracy

Surface reconstruction for micro-samples with large discontinuities using digital holography is a challenge. To overcome this problem, multi-incidence digital holographic profilometry (MIDHP) has been proposed. MIDHP relies on the numerical generation of the longitudinal scanning function (LSF) for reconstructing the topography of the sample with large depth and high axial resolution. Nevertheless, the method is unable to reconstruct surfaces with large gradients due to the need of: (i) high precision focusing that manual adjustment cannot fulfill and (ii) preserving the functionality of the LSF that requires capturing and processing many digital holograms. In this work, we propose a novel MIDHP method to solve these limitations. First, an autofocusing algorithm based on the comparison of shapes obtained by the LSF and the thin tilted element approximation is proposed. It is proven that this autofocusing algorithm is capable to deliver in-focus plane localization with submicron resolution. Second, we propose that wavefield summation for the generation of the LSF is carried out in Fourier space. It is shown that this scheme enables a significant reduction of arithmetic operations and can minimize the number of Fourier transforms needed. Hence, a fast generation of the LSF is possible without compromising its accuracy. The functionality of MIDHP for measuring surfaces with large gradients is supported by numerical and experimental results.

Multifeature Metric Learning Based on Enhanced Equidistance Embedding for Electroencephalogram Recognition of Epilepsy

Mobile edge computing (MEC) has the ability of pattern recognition and intelligent processing of real-time data. Electroencephalogram (EEG) is a very important tool in the study of epilepsy. It provides rich information that can not be provided by other physiological methods. In the automatic classification of EEG signals by intelligent algorithms, feature extraction and the establishment of classifiers are both very important steps. Different feature extraction methods, such as time domain, frequency domain, and nonlinear dynamic feature methods, contain independent and diverse specific information. Using multiple forms of features at the same time can improve the accuracy of epilepsy recognition. In this paper, we apply metric learning to epileptic EEG signal recognition. Inspired by the equidistance constrained metric learning algorithm, we propose multifeature metric learning based on enhanced equidistance embedding (MMLE3) for EEG recognition of epilepsy. The MMLE3 algorithm makes use of various forms of EEG features, and the feature weights are adaptively weighted. It is a big advantage that the feature weight vector can be adjusted adaptively, without manual adjustment. The MMLE3 algorithm maximizes the distance between the samples constrained by the cannot-link, and the samples of different classes are transformed into equidistant; meanwhile, MMLE3 minimizes the distance between the data constrained by the must-link, and the samples of the same class are compressed to one point. Under the premise that the various feature classification tasks are consistent, MMLE3 can fully extract the associated and complementary information hidden between the features. The experimental results on the CHB-MIT dataset verify that the MMLE3 algorithm has good generalization performance.

Adaptive multilayer management remotely operated underwater vehicle

В статье рассматривается самоорганизующаяся адаптивная система управления телеуправляемыми необитаемыми подводными аппаратами. Адаптивная нейронная система многослойного управления построена по принципу декомпозиции мультичастотного набора входных сигналов, формируемых в адаптивном джойстике управления. Декомпозиция наборов последовательностей управляющих сигналов проходит процедуру трешолдинга, разделения по оценкам спектра мультичастотного сигнала управления. Каскадный алгоритм построен по принципу интерполяции и децимации коэффициентов фильтра. Трешолдинг реализуется свёрткой форматного кадра управляющего сигнала с коэффициентами всплеск формирующего фильтра в базисе всплесков Добеши. Интерполяция коэффициентов фильтра происходит сдвигом частоты, децимация схлопыванием коэффициентов фильтра. Спектральные оценки, построенные по среднеквадратическому значению спектра, укладываются в спектральный радиус нормированного сигнала и формируют матрицу математического ожидания адаптивного сигнала управления. Реакции пилота телеуправляемого необитаемого подводного аппарата формируют управляющие сигналы в трёх плоскостях с заданными скоростями и моментами. Трешолдинг в базисе всплесков позволяет формировать сигналы управления с оптимальной крутизной выходной характеристики, что позволяет отказаться от необходимой ручной регулировки мощностей движителей двигательно-рулевого комплекса, при реализации полуавтоматичеcкого и автоматического управления. Обратная связь системы управления по наблюдаемой динамике позволяет реализовать функцию автопилота, с учётом заданных критериев качества. The article discusses the self-organizing adaptive system management remotely operated underwater vehicle. Adaptive neural system of multilayer control is built on the principle of decomposition of multi-frequency set of input signals generated in adaptive joystick of management. The decomposition of the sets of control signal sequences undergoes the procedure of tresholding, separation by estimates of the spectrum of the multi-frequency control signal. The cascade algorithm is based on the principle of interpolation and decimation of filter coefficients. Tresholding is implemented by convolving the format frame of the control signal with wavelet coefficients of the forming filter in the basis of Dobeshi wavelet. Interpolation of filter coefficients occurs by frequency shift, decimation by collapse of filter coefficients. Spectral estimates based on the standard value of the spectrum fit into the spectral radius of the normalized signal and form a matrix of mathematical expectation of the adaptive control signal. The reactions of the pilot of a remotely operated underwater vehicle form control signals in three planes with given speeds and moments. Tresholding in the basis of wavelets allows you to generate control signals with an optimal slope of the output characteristic, which allows you to abandon the necessary manual adjustment of the powers of the propulsion engines of the engine-steering system, when implementing semi-automatic and automatic control. Feedback of the control system according to the observed dynamics allows implementing the autopilot function, taking into account the specified quality criteria.

Interpretable network propagation with application to expanding the repertoire of human proteins that interact with SARS-CoV-2

Background Network propagation has been widely used for nearly 20 years to predict gene functions and phenotypes. Despite the popularity of this approach, little attention has been paid to the question of provenance tracing in this context, e.g., determining how much any experimental observation in the input contributes to the score of every prediction. Results We design a network propagation framework with 2 novel components and apply it to predict human proteins that directly or indirectly interact with SARS-CoV-2 proteins. First, we trace the provenance of each prediction to its experimentally validated sources, which in our case are human proteins experimentally determined to interact with viral proteins. Second, we design a technique that helps to reduce the manual adjustment of parameters by users. We find that for every top-ranking prediction, the highest contribution to its score arises from a direct neighbor in a human protein-protein interaction network. We further analyze these results to develop functional insights on SARS-CoV-2 that expand on known biology such as the connection between endoplasmic reticulum stress, HSPA5, and anti-clotting agents. Conclusions We examine how our provenance-tracing method can be generalized to a broad class of network-based algorithms. We provide a useful resource for the SARS-CoV-2 community that implicates many previously undocumented proteins with putative functional relationships to viral infection. This resource includes potential drugs that can be opportunistically repositioned to target these proteins. We also discuss how our overall framework can be extended to other, newly emerging viruses.

Post-Alignment Adjustment and Its Automation

Multiple sequence alignment (MSA) is the basis for almost all sequence comparison and molecular phylogenetic inferences. Large-scale genomic analyses are typically associated with automated progressive MSA without subsequent manual adjustment, which itself is often error-prone because of the lack of a consistent and explicit criterion. Here, I outlined several commonly encountered alignment errors that cannot be avoided by progressive MSA for nucleotide, amino acid, and codon sequences. Methods that could be automated to fix such alignment errors were then presented. I emphasized the utility of position weight matrix as a new tool for MSA refinement and illustrated its usage by refining the MSA of nucleotide and amino acid sequences. The main advantages of the position weight matrix approach include (1) its use of information from all sequences, in contrast to other commonly used methods based on pairwise alignment scores and inconsistency measures, and (2) its speedy computation, making it suitable for a large number of long viral genomic sequences.

Image-Free Ultrasound Blood-Flow Monitoring Circuit System with Automatic Range-Gate Positioning Scheme: A Pilot Study

This work proposes a proof-of-concept ultrasound blood-flow-monitoring circuit system using a single-element transducer. The circuit system consists of a single-element ultrasonic transducer, an analog interface circuit, and a field-programmable gate array (FPGA). Since the system uses a single-element transducer, an ultrasound image cannot be reconstructed unless scanning with mechanical movement is used. An ultrasound blood-flow monitor basically needs to acquire a Doppler sample volume by positioning a range gate at a vessel region on a scanline. Most recent single-transducer-based ultrasound pulsed-wave Doppler devices rely on a manual adjustment of the range gate to acquire Doppler sample volumes. However, the manual adjustment of the range gate depends on the user’s experience, and it can be time consuming if a transducer is not properly positioned. Thus, automatic range-gate-positioning is more desirable for image-free pulsed-wave Doppler devices. This work proposes a circuit system which includes a new automatic range-gate-positioning scheme. It blindly tracks the position of a blood vessel on a scanline by using the accumulation of Doppler amplitude deviations and a hysteresis slicing function. The proposed range-gate-positioning scheme has been implemented in an FPGA for real-time operation and is based on addition-only computations, except for filter parts to reduce the complexity of computation in the hardware. The proposed blood-flow-monitoring circuit system has been implemented with discrete commercial chips for proof-of-concept purposes. It uses a center frequency of 2 MHz and a system-clock frequency of 20 MHz. The FPGA only utilizes 5.6% of slice look-up-tables (LUTs) for implementation of the range-gate-positioning scheme. For measurements, the circuit system was utilized to interrogate a customized flow phantom model, which included two vessel-mimicking channels. The circuit system successfully acquired Doppler sample volumes by positioning a range gate on a fluid channel. In addition, the estimated Doppler shift frequency shows a good agreement with the theoretical value.

Self-Adaptive Particle Swarm Optimization-Based Echo State Network for Time Series Prediction

Echo state networks (ESNs), belonging to the family of recurrent neural networks (RNNs), are suitable for addressing complex nonlinear tasks due to their rich dynamic characteristics and easy implementation. The reservoir of the ESN is composed of a large number of sparsely connected neurons with randomly generated weight matrices. How to set the structural parameters of the ESN becomes a difficult problem in practical applications. Traditionally, the design of the parameters of the ESN structure is performed manually. The manual adjustment of the ESN parameters is not convenient since it is an extremely challenging and time-consuming task. This paper proposes an ensemble of five particle swarm optimization (PSO) strategies to design the structure of ESN and then reduce the manual intervention in the design process. An adaptive selection mechanism is used for each particle in the evolution to select a strategy from the strategy candidate pool for evolution. In addition, leaky integration neurons are used as reservoir internal neurons, which are added within the adaptive mechanism for optimization. The root mean squared error (RMSE) is adopted as the evaluation criterion. The experimental results on Mackey–Glass time series benchmark dataset show that the proposed method outperforms other traditional evolutionary methods. Furthermore, experimental results on electrocardiogram dataset show that the proposed method on the ensemble of PSO displays an excellent performance on real-world problems.

Enhanced super-resolution microscopy by extreme value based emitter recovery

Super-resolution localization microscopy allows visualization of biological structure at nanoscale resolution. However, the presence of heterogeneous background can degrade the nanoscale resolution by tens of nanometers and introduce significant image artifacts. Here we investigate and validate an efficient approach, referred to as extreme value-based emitter recovery (EVER), to accurately recover the distorted fluorescent emitters from heterogeneous background. Through numerical simulation and biological experiments, we validated the accuracy of EVER in improving the fidelity of the reconstructed super-resolution image for a wide variety of imaging characteristics. EVER requires no manual adjustment of parameters and has been implemented as an easy-to-use ImageJ plugin that can immediately enhance the quality of reconstructed super-resolution images. This method is validated as an efficient way for robust nanoscale imaging of samples with heterogeneous background fluorescence, such as thicker tissue and cells.