Search

An intelligent cleaning vehicle improves the efficiency of road cleaning to a great extent. In this case, road garbage recognition is fundamental and crucial. Usually, the pebbles are too small, and the sand has an inconspicuous boundary and unfixed shape on the roads. These issues make the stones and sand too hard to be detected by a garbage recognition system. Hence, we designed a novel semantic segmentation network that acquires the areas and categories of road garbage. We designed a Deep Supervision and High Fusion (DSHF) block to improve the road garbage segmentation accuracy. The designed block with a backbone network of HFCN and UNet++ is comparable with a model that only has either a deep supervision block or a high fusion block. According to the collected road garbage segmentation dataset comprising four categories (stones, leaves, sand and bottles), the model we designed has improved the metric value of MPA by 3% over the state-of-art methods and achieved the highest MIoU (Mean Intersection over Union) with the value of 77.92%. The results of our experiments show that the semantic segmentation of road garbage is feasible and that the proposed DSHF block is practical for improving the segmentation effect of road garbage.

The design fabrication and cold test of a mode generator to excite the circular TE₀₁ or TE₂₁ mode for H band 220 325 GHz gyrotron traveling wave tubes TWTs are presented in this article The mode generator is composed of two parts: a TE $^{{□ }}_{10}$ -TE $^{□}_{20}$ mode converter based on a two-way H-T junction power divider and a TE $^{□}_{20}$ -TE°₀₁/TE $°_{{21}}$ mode converter with a tapered waveguide (□ and ° represent the rectangular and circular waveguide, respectively). By adjusting the ratio of width and height of the tapered waveguide, the TE°₀₁ or TE $°_{21}$ mode could be excited. The primary advantage of this technique is that it has a relatively compact and easy-to-fabricate configuration, which is especially useful in the terahertz-band. To verify the design, two identical mode generators joined back-to-back were manufactured and measured. The measured S-parameters were in good agreement with the simulated ones, which predicted that the TE $°_{21}$ /TE°₀₁ mode generator obtained a 1-dB transmission bandwidth of ~38.1 and ~48.0 GHz, respectively. Meanwhile the reflection coefficient $S_{11}$ was below -10 dB over that bandwidth.

The dispersion and dielectric attenuation characteristics of a double-staggered grating waveguide (DSGW) are two crucial features related to the output characteristics of a sheet-beam traveling-wave amplifier and have been previously studied by using theoretical analysis and simulations. In this work, these two characteristics are both experimentally studied. Two ${H}$ -band (220–325 GHz) DSGW circuit prototypes with and without dielectric attenuators made of the BeO-TiO2 composite ceramic were machined by using nanocomputer numerical control milling. Experimental studies of the dispersion property were performed successfully where the measurement results were in good agreement with those predicted by simulations, exhibiting a wide passband of 70 GHz (230–300 GHz). The measured transmission coefficient has been decreased from −5 to −55 dB after the introduction of dielectric attenuators, experimentally verifying the feasibility of the adopted methodology. The effects of the variations of dispersion relation and transmission loss on the beam-wave interaction stability were also investigated.

Road garbage recognition and cleanliness assessment are very important for intelligent cleaning of vehicles. However, widely used garbage detection methods cannot provide accurate information to assess cleanliness. Also, only a few contributions are available in the road cleanliness assessment, and they too mainly depend on the category and quantity of garbage. In this paper, deep supervision UNet++ (DUNet++) is proposed to solve the problem of road garbage classification and segmentation, which can directly impact the output of the category of garbage and the occupied ground area. A new, simple and accurate method is established to assess road cleanliness by combining the results of garbage segmentation. Compared with the cleanliness index (CI), the road cleanliness index based on semantic segmentation (CISS) not only considers the road area occupied by different types or categories of the garbage but also considers the difficulty of cleaning them. A road garbage segmentation dataset, which consists of four categories (stones, leaves, sand, and bottles), is collected to train the designed garbage segmentation model, and it has achieved considerable garbage segmentation improvement with an MIoU (Mean Intersection over Union) of $76.73\pm 0.11$ . Especially when compared with the state-of-the-art methods, the model we designed has greatly improved the accuracy of garbage segmentation. Experimental results have shown that our road garbage segmentation and cleanliness assessment methods are relatively simple, and can provide accurate road cleanliness information for cleaning vehicles.

Due to distortion, limitations of vision, and occlusion, most of the existing vacant parking slot detection methods with a standalone around view monitor (AVM) are prone to miss some parking slots and incorrectly identify whether the parking slot is vacant. To overcome this problem, we propose a complete method for vacant parking slot detection and tracking during driving and parking. Considering the different conditions of driving and parking, two different deep convolutional neural networks (DCNNs) are used to detect parking slots, of which the vacant parking slot detection network (VPS-Net) is used to detect vacant parking slots during driving, and the directional marking point detection network (DMPR-PS) is used to detect the directional marking points of the target parking slot during parking. Furthermore, in the driving process, we design a new matching rule and tracking management rule based on the Kernelized Correlation Filter (KCF) to track the parking slots, and fuse classification results of multiple frames to determine the occupancy status. In the parking process, since the parking slot is easily blocked by the vehicle, we design another new tracker to track the directional marking points and infer the complete parking slot using tracking results and prior geometric information. To evaluate the proposed method, a labeled video sequence dataset is established. Experiments show that the proposed method has improved the accuracy and continuity of vacant parking slots detection and positioning whether in the driving process or parking process.

Left-behind humans inside the car or bus have caused a lot of accidents, so it is essential to detect the humans in vehicle. Current human detection methods rely on wearable devices, oxygen sensors, and special seat designs in vehicles, but those sensors cannot adapt to ever-changing environments. To solve those problems and especially to improve passengers’ safety on the bus, we propose a method to accomplishing human detection by fusion vision and microwave radar information in various environments in vehicle. For vision information, we use different networks to extract human and human face features, and fusion of the detection results in different models to improve human detection accuracy. The human detection model is MobileNet-V2, and the human face detection model is MTCNN. A new matching schedule and tracking objects management rule based on the Kernelized Correlation Filter tracker are designed to track the human and human face detection boxes. The microwave radar information is used to detect moving objects. Finally, the fusion vision and microwave radar detection results are implemented. Experiments show that our method has improved the human detection accuracy in vehicle, and this method can be used for detection of left-behind children on the school bus.

A W-band TE□ 40 mode converter with a compact structure based on three T-junction power dividers for measuring high-order overmoded mode slow-wave structures is presented. Since the T-junction discontinuity causes the port reflection, a series of matching steps were introduced in the converter. The proposed converter possessed a bandwidth of 12 GHz with the transmission coefficient S 21 higher than -3 dB and port reflection S 11 lower than -15 dB. More importantly, the transmission coefficient of the unwanted modes is lower than -15 dB within the bandwidth of interest.

The methodology of using the comibination of high-order overmoded slow-wave structure (SWS) and multiple sheet electron beams to increase the output power of traveling-wave tubes or backward-wave oscillators operating in the terahertz-band is presented in this paper. In this paper, recent studies regarding high-order overmoded (HOOM) SWS based multiple sheet electron beam devices at the Shenzhen University have been presented. To realize such an idea, we have so far proposed several kinds of novel HOOM SWS and HOOM power coupling structures. Simulation and measurement results demonstrated that they have achieved good performace.

Preliminary study of a TE20 input/output coupling structure for sheet electron beam traveling wave tubes is presented in this paper. A TE10-TE20 mode converter and a single branch waveguide coupler are used to realize the proposed design. Simulation results obtained by using the CST simulation software demonstrated that the port reflection and the transmission coefficient were respectively lower than - 15 dB and higher than - 1 dB in the frequency range of 241-298 GHz. The port isolation was lower than - 10 dB in the frequency range of 250-298 GHz.

The automated valet parking (AVP) based on the around view monitor (AVM) has attracted increasing attention from the research community due to its high practicality and low cost. However, the distortion correction and inverse perspective transformation of the around view images result in the truncated and deformed objects, which leads to poor detection performance. To address this problem, we propose a method of object ground lines regression and mapping from fisheye images to around view images. There are two key steps for the proposed method, regressing ground lines in the fisheye images and mapping ground lines to the around view image. To detect the objects and regress ground lines at the same time, we explored three new detectors based on YOLOv5 (i.e. YOLOv5-VGG, YOLOv5-Point, and YOLOv5-Ratio). Among them, YOLOv5-VGG adds a VGG-based regression network behind YOLOv5, YOLOv5-Point directly adds a regression head to share the convolutional features, and YOLOv5-Ratio combines the 2D box and ground line to predict its ratios. After that, the ground lines from four different views are mapped and fused to the around view image. Additionally, to facilitate the study of the object detection in the around view image, a large-scale labeled dataset is established, which comprises 9828 fisheye images collected from typical indoor and outdoor parking lots. For each image, the 2D bounding boxes and ground lines of objects are carefully labeled. Experiments show that the YOLOv5-Ratio obtains the best performance and can effectively detect the truncated and deformed object in the around view image with the precision rate of 92.28% and recall rate of 85.09% on our collected dataset.

The design and simulation of a terahertz high-order mode power coupler for sheet beam traveling wave amplifiers are presented in this paper. The proposed power coupler has functions of both the mode conversion and the beam wave convergence/separation. To improve the performance of this structure, some auxiliary structures are introduced, including a rectangular septum, a matching step and a pair of symmetric Bragg reflectors. The proposed power coupler has numerous competitive advantages, such as broad bandwidth, high transmission coefficient and high conversion efficiency, high port isolation, low port reflection, and high mode purity.

The sheet beam travelling wave tube (SB-TWT) is a promising sub-terahertz amplifier with high output power and wide frequency bandwidth. The design of key components operating within the H-band (220–325 GHz), including a sheet electron beam gun, an offset-pole periodic cusped magnet, and a high frequency system composed of input/output couplers as well as a double staggered grating waveguide slow-wave structure, is presented in this paper. Beam-wave interaction simulations with the ohmic loss being taken into account predicted that the amplifier produced an output power of >55 W in the frequency range of 230–280 GHz.

Linear feature extraction is crucial for special objects in semantic segmentation networks, such as slot marking and lanes. The objects with linear characteristics have global contextual information dependency. It is very difficult to capture the complete information of these objects in semantic segmentation tasks. To improve the linear feature extraction ability of the semantic segmentation network, we propose introducing the dilated convolution with vertical and horizontal kernels (DVH) into the task of feature extraction in semantic segmentation networks. Meanwhile, we figure out the outcome if we put the different vertical and horizontal kernels on different places in the semantic segmentation networks. Our networks are trained on the basis of the SS dataset, the TuSimple lane dataset and the Massachusetts Roads dataset. These datasets consist of slot marking, lanes, and road images. The research results show that our method improves the accuracy of the slot marking segmentation of the SS dataset by 2%. Compared with other state-of-the-art methods, our UnetDVH-Linear (v1) obtains better accuracy on the TuSimple Benchmark Lane Detection Challenge with a value of 97.53%. To prove the generalization of our models, road segmentation experiments were performed on aerial images. Without data argumentation, the segmentation accuracy of our model on the Massachusetts roads dataset is 95.3%. Moreover, our models perform better than other models when training with the same loss function and experimental settings. The experiment result shows that the dilated convolution with vertical and horizontal kernels will enhance the neural network on linear feature extraction.

Background Effective postoperative analgesia may enhance early rehabilitation after total knee arthroplasty (TKA). The purpose of this study is to perform a randomized controlled trial to compare the efficiency of adductor canal block (ACB) with periarticular infiltration (PAI) versus PAI alone for early postoperative pain treatment after TKA. Methods After institutional review board approval, written informed consent was obtained from patients undergoing elective TKA. Subjects were randomized into 2 groups as follows: adductor canal blockade with 30 mL of 0.5% ropivacaine and 100 mcg of clonidine. All patients received a periarticular infiltration mixture intraoperatively with scheduled and patient requested oral and IV analgesics postoperatively for breakthrough pain. The primary outcome was morphine consumption in the first 24 hours. Secondary outcomes included pain scores, morphine consumption at 48 hours, opioid-related side effects (post-operative nausea/vomiting, sedation scores), functional outcomes, quadriceps strength, and length of hospital stay. Conclusions For the present trial, we hypothesized that patients receiving adductor canal block + PAI would have significantly lower morphine consumption and pain scores after surgery. Trial registration number researchregistry5490.