Moving vehicle detection, tracking, and counting are very critical for traffic flow monitoring, planning, controlling. Video-based solution, comparing to other techniques, does not disturb is easily installed. By analyzing the video sequence recorded from a camera, this paper presents video-based solution applied with adaptive subtracted background technology in combination virtual detector blob tracking technologies. Experimental results, implemented Visual C++ code OpenCV development kits,...

The capability to simultaneously interfere with multiple hostile targets is of great importance for modern electronic jammers. To allow the radio frequency amplifier operate at maximum efficiency and avoid nonlinearity in components, low peak-to-average-power-ratio (PAPR) multi-target jamming signals are desired. In this paper, we focus on design low-PAPR a jammer digital arrays. end, propose optimize amplitude phase each component signals. By virtue an approximation method, show that...

This paper focuses on the joint design of transmit waveforms and receive filters for airborne multiple-input-multiple-output (MIMO) radar systems in spectrally crowded environments. The purpose is to maximize output signal-to-interference-plus-noise-ratio (SINR) presence signal-dependent clutter. To improve practicability waveforms, both a multi-spectral constraint peak-to-average-power ratio (PAPR) are imposed. A cyclic method derived iteratively optimize filters. In particular, tackle...

Meta-learning and other approaches to few-shot learning are widely studied for image recognition, increasingly applied vision tasks such as pose estimation dense prediction. This naturally raises the question of whether there is any fewshot metalearning algorithm capable generalizing across these diverse task types? To support community in answering this question, we introduce Meta Omnium, a dataset-of-datasets spanning multiple including keypoint localization, semantic segmentation...

Fingerprint-based positioning techniques are a hot research topic because of their satisfactory accuracy in complex environments. In this study, we adopted the deep-learning-based long-time-evolution (LTE) signal fingerprint method for outdoor environment positioning. Inspired by state-of-the-art image classification methods, novel hybrid location gray-scale utilizing LTE fingerprints is proposed paper. order to deal with fluctuations, several data enhancement methods adopted. A hierarchical...

Wi-Fi and magnetic field fingerprinting-based localization have gained increased attention owing to their satisfactory accuracy global availability. The common signal-based fingerprint deteriorates due well-known signal fluctuations. In this paper, we proposed a field-based system based on deep learning. Owing the low discernibility of strength (MFS) in large areas, unsupervised learning density peak clustering algorithm comparison distance (CDPC) is first used pick up several center points...

We have implemented a single-photon detection system in the 1.55-μm region, based on frequency up-conversion MgO-doped periodically poled LiNbO3 waveguide. subsequently reached record-low dark count rate of 45 counts per second. The detectable signal photon rate, i.e., for counting up-converted photons corrected by counts, noises, and losses, reaches value 81 Through free-space coupling, we eliminated induced parametric fluorescence.

In this article, we consider waveform design for dual-function radar communication (DFRC) systems based on multiple-input–multiple-out (MIMO) arrays. To achieve better Rician target detection performance, use the relative entropy associated with formulated problem as metric. We also impose a multiuser interference (MUI) energy constraint waveforms to ensure achievable sum rate of communications. Two algorithms are presented tackle nonlinear nonconvex problem. first algorithm, derive...

In order to address the data security and communication efficiency of vehicles during high-speed mobile communication, this paper investigates problem secure in-vehicle resource allocation based on slow-variable large-scale fading channel information, meet quality service requirements vehicular i.e., ensure reliability V2V time delay while maximizing transmission rate cellular link. And an eavesdropping model is introduced delivery link information. Considering that high mobility causes...

A switchable triple-wavelength thulium-doped fibre laser based on an all-fibre Mach–Zehnder interferometer and ring filter with a polarization-maintaining is proposed experimentally demonstrated. In the laser, comprising two 1 × 2 optical couplers inserted into cavity to produce comb effect. The comprises 3:7 splitting ratio 2-m-long improve lasing stability. Single-wavelength can be tuned continuously from 1864.4 1884.5 nm, five different modes of dual-wavelength lasers realized by changing...

Intelligent Reflective Surfaces (IRS), as a passive, low-cost, and green technology, are widely considered to address the high complexity, hardware costs, increased power consumption pressures faced by future wireless networks provide large-scale connectivity services larger number of devices around world. This paper summarizes advantages IRS technology compared with other technologies, demonstrates effectiveness assisted communication in new scenarios combined related technologies. In...

In the process of Chinese license plate recognition, main problems are as follows, such feature extraction method is cumbersome and inefficient, character recognition rate low. This paper studies technology in depth based on convolution neural network. order to solve problem low characters, we use continuous layers convolve image extract more characters. Experiments show that proposed can effectively characteristics, improve rate.

An octagonal FPSO has been proposed for marginal oil and gas development in shallow waters. A shuttle tanker will be deployed near the during offloading operations. This new concept simplifies design manufacturing processes, yet maintains full production, storage, functions of a conventional ship-shaped FPSO. However, mooring system this floating unit imposes technical challenges due to: 1) high environmental loads expected on unit, 2) large dynamic offsets waters, 3) inadequate performance...

Spreadsheets are widely used but subject to various defects. In this paper, we present SGuard effectively detect spreadsheet learns features cluster cells with similar computational semantics, and then refines these clusters recognize anomalous as well balances the trade-off between precision (87.8%) recall rate (71.9%) in defect detection, achieves an F-measure of 0.79, exceeding existing detection techniques. We introduce implementation its usage by a video presentation...

Localization has attracted considerable attention in recent years because of the growth location-based services and computer-aided technology. In particular, wireless fingerprint-based positioning techniques have received due to their satisfactory performance with smartphones. this work, a two-level hierarchical structure system is proposed achieve accuracy. First, long-time-evolution (LTE) mobile networks are used provide information, collected measurements LTE signals converted into...

Two methods are adopted to remove wetting agents on surface of the quartz fiber, one is heat treatment, and other combination organic solvent treatment. By comparing changes weight loss rate fiber before after analyzing variation morphology composition in SEM XPS approaches, we conclude that treatment superior pure thermal dealing with fiber. In heating process, different rates used, results show that: lower is, greater mechanical strength losses.

Human pose estimation technology has important applications in many fields. Existing research mainly focuses on precise localization of human key points unobstructed situations but neglects the common occlusion problem daily environments. Occlusion can disrupt correlation between body, causing a decrease model accuracy. In response to above issues, this article proposes joint data augmentation and attention mechanism for occluded estimation. This training images, generating specific number...