scholarly journals Dense-RefineDet for Traffic Sign Detection and Classification

Sensors ◽  
2020 ◽  
Vol 20 (22) ◽  
pp. 6570
Author(s):  
Chang Sun ◽  
Yibo Ai ◽  
Sheng Wang ◽  
Weidong Zhang
Keyword(s):  
High Speed ◽  
Large Scale ◽  
Design Method ◽  
Contextual Information ◽  
Real Life ◽  
Single Shot ◽  
Traffic Sign ◽  
Traffic Signs ◽  
High Level ◽  
Anchor Design

Detecting and classifying real-life small traffic signs from large input images is difficult due to their occupying fewer pixels relative to larger targets. To address this challenge, we proposed a deep-learning-based model (Dense-RefineDet) that applies a single-shot, object-detection framework (RefineDet) to maintain a suitable accuracy–speed trade-off. We constructed a dense connection-related transfer-connection block to combine high-level feature layers with low-level feature layers to optimize the use of the higher layers to obtain additional contextual information. Additionally, we presented an anchor-design method to provide suitable anchors for detecting small traffic signs. Experiments using the Tsinghua-Tencent 100K dataset demonstrated that Dense-RefineDet achieved competitive accuracy at high-speed detection (0.13 s/frame) of small-, medium-, and large-scale traffic signs (recall: 84.3%, 95.2%, and 92.6%; precision: 83.9%, 95.6%, and 94.0%). Moreover, experiments using the Caltech pedestrian dataset indicated that the miss rate of Dense-RefineDet was 54.03% (pedestrian height > 20 pixels), which outperformed other state-of-the-art methods.

scholarly journals Automated Visual Recognizability Evaluation of Traffic Sign Based on 3D LiDAR Point Clouds

2019 ◽  
Vol 11 (12) ◽  
pp. 1453 ◽  
Author(s):  
Shanxin Zhang ◽  
Cheng Wang ◽  
Lili Lin ◽  
Chenglu Wen ◽  
Chenhui Yang ◽  
...  
Keyword(s):  
Traffic Safety ◽  
Laser Scanning ◽  
Large Scale ◽  
Evaluation Method ◽  
Evaluation Model ◽  
Point Clouds ◽  
Traffic Sign ◽  
Traffic Signs ◽  
3D Point Clouds ◽  
Traffic Environment

Maintaining the high visual recognizability of traffic signs for traffic safety is a key matter for road network management. Mobile Laser Scanning (MLS) systems provide efficient way of 3D measurement over large-scale traffic environment. This paper presents a quantitative visual recognizability evaluation method for traffic signs in large-scale traffic environment based on traffic recognition theory and MLS 3D point clouds. We first propose the Visibility Evaluation Model (VEM) to quantitatively describe the visibility of traffic sign from any given viewpoint, then we proposed the concept of visual recognizability field and Traffic Sign Visual Recognizability Evaluation Model (TSVREM) to measure the visual recognizability of a traffic sign. Finally, we present an automatic TSVREM calculation algorithm for MLS 3D point clouds. Experimental results on real MLS 3D point clouds show that the proposed method is feasible and efficient.

A reconfigurable membrane-type acoustic metasurface for low-frequency and broadband wave front modulation

2019 ◽  
Vol 33 (19) ◽  
pp. 1950208
Author(s):  
Xinpei Song ◽  
Tianning Chen ◽  
Jian Zhu ◽  
Yanbin He
Keyword(s):  
Wave Front ◽  
High Speed ◽  
Power Transmission ◽  
Design Method ◽  
Real Life ◽  
Low Frequency ◽  
Pumping System ◽  
Unit Cells ◽  
Membrane Type ◽  
Structure Complexity

Low-frequency and broadband are the critical challenges in real-life applications. Here, we try to tackle the challenges by proposing a reconfigurable acoustic metasurface (AM) composed of the membrane-type metamaterial (MAM) structure of deep sub-wavelength scale. By employing the external air pumping system into each individual unit cell of the AM, the tension of the membrane can be readily tailored by the system with little interference from other unit cells. Two strategies of the constant pressure method (CPM) and constant volume method (CVM) are reported to design the MAM. And the CVM is adopted as the ultimate design strategy by comparing both methods from aspects of the dimension, operating frequency, and structure complexity. In order to validate the low-frequency and broadband performances of the AM, the Airy-like beams and the acoustic converging based on two identical Airy-like beams are introduced and proof-of-concept simulations are performed with the finite element method. The simulated results agree well with the theoretical predictions. Our design provides the little-interference active design method for the low-frequency and broadband AM to manipulate the wave front, and may have practical engineering applications in areas of the aerospace, high-speed train, marine vessel, and power transmission and transformation project.

scholarly journals High-Speed Railway Clearance Intrusion Detection with Improved SSD Network

2019 ◽  
Vol 9 (15) ◽  
pp. 2981 ◽  
Author(s):  
Baoqing Guo ◽  
Jiafeng Shi ◽  
Liqiang Zhu ◽  
Zujun Yu
Keyword(s):  
Intrusion Detection ◽  
High Speed ◽  
Rapid Development ◽  
Single Shot ◽  
L1 Norm ◽  
Low Level ◽  
Model Size ◽  
Public Dataset ◽  
The Mean ◽  
High Level

With the rapid development of high-speed railways, any objects intruding railway clearance will do great threat to railway operations. Accurate and effective intrusion detection is very important. An original Single Shot multibox Detector (SSD) can be used to detect intruding objects except small ones. In this paper, high-level features are deconvolved to low-level and fused with original low-level features to enhance their semantic information. By this way, the mean average precision (mAP) of the improved SSD algorithm is increased. In order to decrease the parameters of the improved SSD network, the L1 norm of convolution kernel is used to prune the network. Under this criterion, both the model size and calculation load are greatly reduced within the permitted precision loss. Experiments show that the mAP of our method on PASCAL VOC public dataset and our railway datasets have increased by 2.52% and 4.74% respectively, when compared to the original SSD. With our method, the elapsed time of each frame is only 31 ms on GeForce GTX1060.

Dynamic Scheduling Model and Algorithm for Dispatching Trains in High-Speed Railway Passenger Station

2010 ◽  
Author(s):  
Yixiang Yue ◽  
Leishan Zhou
Keyword(s):  
High Speed ◽  
Large Scale ◽  
Job Shop ◽  
Dynamic Scheduling ◽  
Time Window ◽  
Real Life ◽  
Greedy Heuristic ◽  
Railway Station ◽  
High Speed Railway ◽  
Railway Passenger

Regarding the railway station tracks and train running routes as machines, all trains in this railway station as jobs, dispatching trains in high-speed railway passenger stations can be considered as a special type of Job-Shop Problem (JSP). In this paper, we proposed a multi-machines, multi-jobs JSP model with special constraints for Operation Plan Scheduling Problem (OPSP) in high-speed railway passenger stations, and presented a fast heuristic algorithm based on greedy heuristic. This algorithm first divided all operations into several layers according to the yards attributes and the operation’s urgency level. Then every operation was allotted a feasible time window, each operation was assigned to a specified “machine” sequenced or backward sequenced within the time slot, layer by layer according to its priority. As we recorded and modified the time slots dynamically, the searching space was decreased dramatically. And we take the South Beijing High-speed Railway Station as example and give extensive numerical experiment. Computational results based on real-life instance show that the algorithm has significant merits for large scale problems; can both reduce tardiness and shorten cycle times. The empirical evidence also proved that this algorithm is industrial practicable.

scholarly journals Study on the Steepness of Landslide-Induced Wave and Nonlinear Motion of Container Ships

2020 ◽  
Vol 10 (6) ◽  
pp. 2159
Author(s):  
Peiyin Yuan ◽  
Pingyi Wang ◽  
Yu Zhao
Keyword(s):  
High Speed ◽  
Large Scale ◽  
Design Method ◽  
Slope Angle ◽  
Volume Effect ◽  
Water Area ◽  
Three Gorges Reservoir Area ◽  
Orthogonal Experimental Design ◽  
Landslide Volume ◽  
Ship Rolling

Large-scale instability of a landslide body, sliding down a bank slope and entering water at a high speed, arouses landslide surges. Taking the water entry point as a source point, they spread rapidly to surrounding areas, increasing the danger risk of vessel passages in the water area. In this paper, adopting an orthogonal experimental design method, based on the test data, the Three Gorges Reservoir area was derived in order to calculate the height of the first wave of the bank landslide surge: to analyze the slope angle, the geological environment, the volume of the landslide surge, and the landslide surge wave steepness; to study the landslide volume effect on ship rolling and the swaying motion rule; and to explore the landslide surge in different ship rolling positions and transverse oscillation characteristics. This study can provide theoretical support for the navigation safety of ships in landslide surge waters.

Dynamic Analysis and Design Method Study on the Combined-Boom System of Portal Crane

2012 ◽  
Vol 152-154 ◽  
pp. 1645-1649 ◽  
Author(s):  
Yuan Tao Sun ◽  
Duan Li
Keyword(s):  
High Speed ◽  
Large Scale ◽  
Design Method ◽  
Time History ◽  
Analysis Method ◽  
Drive Power ◽  
Analysis And Design ◽  
Whole Process ◽  
Handling Device ◽  
Portal Crane

As a main handling device the portal crane is widely used in port, railroad, etc.The crane handling procedure is mainly carried out through its combined-boom system luffing or swing .In general, in order to reduce drive power and improve the operational performance, the luffing trajectory should meet the design requirement. At the same time, structure stress should be secured in the whole process of handling the cargo. Recently, to deal with more heaver and further cargo, the portal crane is becoming more large-scale. So that the large-scale components such as jib elastic deformation effect on large displacement motion cannot be ignored longer. In addition to the structure high speed motion in the process of handling also make the structure dynamic behaviors spending more obvious specially in the condition of luffing combined with swing. However, the problem for this dynamic behavior brings about to physical design sometimes has no method to solve according to the conventional analysis algorithm and dynamics method. To reduce the deviation caused by the common analysis, design and analysis method based on the multibody is put forward in this thesis. According to the method, the result on the luffing trajectory and stress-time history are analyzed easily. So that it ensure the efficiency and increase the accuracy of the initial design according to the conventional design and analysis method.

scholarly journals SSD7-FFAM: A Real-Time Object Detection Network Friendly to Embedded Devices from Scratch

2021 ◽  
Vol 11 (3) ◽  
pp. 1096
Author(s):  
Qing Li ◽  
Yingcheng Lin ◽  
Wei He
Keyword(s):  
Object Detection ◽  
Real Time ◽  
Large Scale ◽  
Feature Fusion ◽  
Contextual Information ◽  
Attention Mechanism ◽  
Detection Accuracy ◽  
Single Shot ◽  
Feature Maps ◽  
Embedded Devices

The high requirements for computing and memory are the biggest challenges in deploying existing object detection networks to embedded devices. Living lightweight object detectors directly use lightweight neural network architectures such as MobileNet or ShuffleNet pre-trained on large-scale classification datasets, which results in poor network structure flexibility and is not suitable for some specific scenarios. In this paper, we propose a lightweight object detection network Single-Shot MultiBox Detector (SSD)7-Feature Fusion and Attention Mechanism (FFAM), which saves storage space and reduces the amount of calculation by reducing the number of convolutional layers. We offer a novel Feature Fusion and Attention Mechanism (FFAM) method to improve detection accuracy. Firstly, the FFAM method fuses high-level semantic information-rich feature maps with low-level feature maps to improve small objects’ detection accuracy. The lightweight attention mechanism cascaded by channels and spatial attention modules is employed to enhance the target’s contextual information and guide the network to focus on its easy-to-recognize features. The SSD7-FFAM achieves 83.7% mean Average Precision (mAP), 1.66 MB parameters, and 0.033 s average running time on the NWPU VHR-10 dataset. The results indicate that the proposed SSD7-FFAM is more suitable for deployment to embedded devices for real-time object detection.

scholarly journals AUTOMATED VISIBILITY FIELD EVALUATION OF TRAFFIC SIGN BASED ON 3D LIDAR POINT CLOUDS

2019 ◽  
Vol XLII-2/W13 ◽  
pp. 1185-1190
Author(s):  
S. Zhang ◽  
C. Wang ◽  
M. Cheng ◽  
J. Li
Keyword(s):  
Traffic Safety ◽  
Traffic Management ◽  
Laser Scanning ◽  
Large Scale ◽  
Field Evaluation ◽  
Point Clouds ◽  
Human Vision ◽  
3D Registration ◽  
Traffic Sign ◽  
Traffic Signs

<p><strong>Abstract.</strong> Maintaining high visibility of traffic signs is very important for traffic safety. Manual inspection and removal of occlusion in front of traffic signs is one of the daily tasks of the traffic management department. This paper presents a method that can automatically detect the occlusion and continuously quantitative estimate the visibility of traffic sign cover all the road surface based on Mobile Laser Scanning (MLS) systems. The concept of traffic sign’s visibility field is proposed in this paper. One of important innovation of this paper is that we use retinal imaging area to evaluate the visibility of a traffic sign. And this makes our method is in line with human vision. To validate the reasonable and accuracy of our method, we use the 2D and 3D registration technology to observe the consistence of the occlusion ratio in point clouds with it in photo. Experiment of implementation on large scale traffic environments show that our method is feasible and efficient.</p>

scholarly journals The development of real-time communication network based on Zigbee and Ethernet for Photovoltaic systems

2014 ◽  
Vol 17 (1) ◽  
pp. 5-15
Author(s):  
Dung Quoc Phan ◽  
Dat Ngoc Dao ◽  
Hiep Chi Le
Keyword(s):  
Communication Network ◽  
Real Time ◽  
High Speed ◽  
Large Scale ◽  
Low Cost ◽  
Photovoltaic System ◽  
Solar Inverter ◽  
Ethernet Controller ◽  
Controller Board ◽  
High Level

In a large system with a lot of distribution solar sources which are all connected to the national grid, a communication system becomes the important part for data acquisition in order to control the whole system stable and efficiency. To deal with this challenge, this paper presents a solution based on Zigbee and Ethernet communication standard. Zigbee standard was created to be a specification of a high level wireless communication protocol which is not only secure, reliable, simple but also low cost and low power. With Zigbee, we can create a communication network for hundreds to thousands of mini solar sources in a large scale of photovoltaic system. Ethernet is a high speed wired communication technology that is used widely in industrial and automatic applications. Together Zigbee and Ethernet bring to us a real-time communication solution for the system. In the experiment prototype of this paper, we use the CC2530ZNP-Mini Kit to create a simple network includes one coordinate and one end device for the first step. The end device was configured to get current and voltage values from a 3-phase grid-connected solar inverter 800Wpk and then sends the values to the coordinate. After the coordinate received data, it would send them to an Ethernet controller board. To display the data through Ethernet, we embedded a web server on the Ethernet controller board. By this way, the data was easy to visualize and supervised by using any web browser.

scholarly journals FAST DRAWING OF TRAFFIC SIGN USING MOBILE MAPPING SYSTEM

2016 ◽  
Vol XLI-B3 ◽  
pp. 937-944
Author(s):  
Q. Yao ◽  
B. Tan ◽  
Y. Huang
Keyword(s):  
Asset Management ◽  
Large Scale ◽  
Rare Event ◽  
Drawing Process ◽  
Traffic Sign ◽  
Mobile Mapping ◽  
Traffic Signs ◽  
Mapping System ◽  
Mobile Mapping System ◽  
Anchor Points

Traffic sign provides road users with the specified instruction and information to enhance traffic safety. Automatic detection of traffic sign is important for navigation, autonomous driving, transportation asset management, etc. With the advance of laser and imaging sensors, Mobile Mapping System (MMS) becomes widely used in transportation agencies to map the transportation infrastructure. Although many algorithms of traffic sign detection are developed in the literature, they are still a tradeoff between the detection speed and accuracy, especially for the large-scale mobile mapping of both the rural and urban roads. This paper is motivated to efficiently survey traffic signs while mapping the road network and the roadside landscape. Inspired by the manual delineation of traffic sign, a drawing strategy is proposed to quickly approximate the boundary of traffic sign. Both the shape and color prior of the traffic sign are simultaneously involved during the drawing process. The most common speed-limit sign circle and the statistic color model of traffic sign are studied in this paper. Anchor points of traffic sign edge are located with the local maxima of color and gradient difference. Starting with the anchor points, contour of traffic sign is drawn smartly along the most significant direction of color and intensity consistency. The drawing process is also constrained by the curvature feature of the traffic sign circle. The drawing of linear growth is discarded immediately if it fails to form an arc over some steps. The Kalman filter principle is adopted to predict the temporal context of traffic sign. Based on the estimated point,we can predict and double check the traffic sign in consecutive frames.The event probability of having a traffic sign over the consecutive observations is compared with the null hypothesis of no perceptible traffic sign. The temporally salient traffic sign is then detected statistically and automatically as the rare event of having a traffic sign.The proposed algorithm is tested with a diverse set of images that are taken inWuhan, China with theMMS ofWuhan University. Experimental results demonstrate that the proposed algorithm can detect traffic signs at the rate of over 80% in around 10 milliseconds. It is promising for the large-scale traffic sign survey and change detection using the mobile mapping system.

Sign in / Sign up

Export Citation Format

Share Document