scholarly journals Analysis of Vehicle-Following Heterogeneity Using Self-Organizing Feature Maps

2014 ◽  
Vol 2014 ◽  
pp. 1-11 ◽  
Author(s):  
Jie Yang ◽  
Ruey Long Cheu ◽  
Xiucheng Guo ◽  
Alicia Romo
Keyword(s):  
Relative Velocity ◽  
Data Sets ◽  
Feature Maps ◽  
Passenger Cars ◽  
Passenger Car ◽  
Feature Map ◽  
Car Following ◽  
Vehicle Type ◽  
Vehicle Trajectories ◽  
Self Organizing

A self-organizing feature map (SOM) was used to represent vehicle-following and to analyze the heterogeneities in vehicle-following behavior. The SOM was constructed in such a way that the prototype vectors represented vehicle-following stimuli (the follower’s velocity, relative velocity, and gap) while the output signals represented the response (the follower’s acceleration). Vehicle trajectories collected at a northbound segment of Interstate 80 Freeway at Emeryville, CA, were used to train the SOM. The trajectory information of two selected pairs of passenger cars was then fed into the trained SOM to identify similar stimuli experienced by the followers. The observed responses, when the stimuli were classified by the SOM into the same category, were compared to discover the interdriver heterogeneity. The acceleration profile of another passenger car was analyzed in the same fashion to observe the interdriver heterogeneity. The distribution of responses derived from data sets of car-following-car and car-following-truck, respectively, was compared to ascertain inter-vehicle-type heterogeneity.

Does Separating Trucks from Other Traffic Improve Overall Safety?

2005 ◽  
Vol 1922 (1) ◽  
pp. 156-166 ◽  
Author(s):  
Dominique Lord ◽  
Dan Middleton ◽  
Jeffrey Whitacre
Keyword(s):  
Clear Understanding ◽  
Mixed Traffic ◽  
Contributing Factors ◽  
Truck Traffic ◽  
Passenger Cars ◽  
Construction Costs ◽  
Passenger Car ◽  
Good Opportunity ◽  
Crash Data ◽  
Vehicle Type

Decision makers have long speculated that building separate roads for trucks and passenger cars, or at least separating these into their own lanes, would accomplish two major objectives: (a) roadways would be made safer for passenger cars and (b) roadways designed specifically for a select class of vehicles rather than for all vehicles might represent overall savings in construction costs. This paper addresses the first objective. Recent studies on the evaluation of safety effects of truck traffic levels on general freeway facilities have not provided a clear understanding of how they affect the number of crashes. In some cases, studies have been contradictory. In addition, no studies have specifically compared passenger car–only with mixed-traffic freeway facilities. The research on which this paper is based aimed to assess whether more homogeneous flows of traffic by vehicle type are safer than the current mixed-flow scenario. An exploratory analysis of crash data was conducted on selected freeway sections of the New Jersey Turnpike for 2002. These sections operate as a dual–dual freeway facility: divided inner and outer lanes. At these locations, the inner lanes have the special characteristic of being for passenger cars only (homogeneous traffic). The selected sections, therefore, offer a very good opportunity to compare the crash experience between passenger car–only and mixed-traffic rural freeway facilities. The results of the study show that outer lanes experience more crashes, both when raw numbers are used and when exposure is included in the analysis. It was shown that truck-related crashes contribute significantly to the total number of crashes on the outer lanes. In fact, trucks are overinvolved in crashes given the exposure on these sections. Although the outcome of this study suggests that separating truck traffic from passenger cars for freeway facilities improves safety, further work is needed to understand the contributing factors leading to truck-related crashes in the outer lanes.

Controlling the Magnification Factor of Self-Organizing Feature Maps

1996 ◽  
Vol 8 (4) ◽  
pp. 757-771 ◽  
Author(s):  
H.-U. Bauer ◽  
R. Der
Keyword(s):  
Information Transfer ◽  
Categorical Perception ◽  
Feature Maps ◽  
Feature Map ◽  
Magnet Effect ◽  
Perceptual Magnet Effect ◽  
Optimal Value ◽  
Optimal Information ◽  
One Dimensional Maps ◽  
Self Organizing

The magnification exponents μ occurring in adaptive map formation algorithms like Kohonen's self-organizing feature map deviate for the information theoretically optimal value μ = 1 as well as from the values that optimize, e.g., the mean square distortion error (μ = 1/3 for one-dimensional maps). At the same time, models for categorical perception such as the "perceptual magnet" effect, which are based on topographic maps, require negative magnification exponents μ < 0. We present an extension of the self-organizing feature map algorithm, which utilizes adaptive local learning step sizes to actually control the magnification properties of the map. By change of a single parameter, maps with optimal information transfer, with various minimal reconstruction errors, or with an inverted magnification can be generated. Analytic results on this new algorithm are complemented by numerical simulations.

scholarly journals Analyzing the Impact of Trucks on Traffic Flow Based on an Improved Cellular Automaton Model

2016 ◽  
Vol 2016 ◽  
pp. 1-14 ◽  
Author(s):  
Dewen Kong ◽  
Xiucheng Guo ◽  
Bo Yang ◽  
Dingxin Wu
Keyword(s):  
Cellular Automaton ◽  
Traffic Flow ◽  
Traffic Congestion ◽  
Traffic Volume ◽  
Cellular Automaton Model ◽  
Passenger Cars ◽  
Passenger Car ◽  
Car Following ◽  
Proposed Model ◽  
The Impact

This paper aims to analyze the impact of trucks on traffic flow and propose an improved cellular automaton model, which considers both the performance difference between passenger cars and trucks and the behaviour change of passenger cars under the impact of trucks. A questionnaire survey has been conducted to find out whether the impact of trucks exists and how the behaviour of passenger car drivers changes under the impact of trucks. The survey results confirm that the impact of trucks exists and indicate that passenger car drivers will enlarge the space gap, decelerate, and change lanes in advance when they are affected. Simulation results show that traffic volume is still affected by percentages of trucks in the congestion phase in the proposed model compared with traditional heterogeneous cellular automaton models. Traffic volume and speed decrease with the impact of trucks in the congestion phase. The impact of trucks can increase traffic congestion as it increases. However, it has different influences on the speed variance of passenger cars in different occupancies. In the proposed model, the relative relationship of the space gap between car-following-truck and car-following-car is changeable at a certain value of occupancy, which is related to the impact of trucks.

scholarly journals Hyper Spectral Image Classification using Multi Labelled, Multi-Scale and Multi-Angle CNN with MS-MA BT Algorithm

2019 ◽  
Vol 8 (9S3) ◽  
pp. 1229-1234
Keyword(s):  
Neural Networks ◽  
High Performance ◽  
Hyperspectral Image ◽  
Support Vector ◽  
Data Sets ◽  
Feature Maps ◽  
Feature Map ◽  
Convolution Neural Networks ◽  
Multi Scale ◽  
Suggested Technique

For classifying the hyperspectral image (HSI), convolution neural networks are used widely as it gives high performance and better results. For stronger prediction this paper presents new structure that benefit from both MS - MA BT (multi-scale multi-angle breaking ties) and CNN algorithm. We build a new MS - MA BT and CNN architecture. It obtains multiple characteristics from the raw image as an input. This algorithm generates relevant feature maps which are fed into concatenating layer to form combined feature map. The obtained mixed feature map is then placed into the subsequent stages to estimate the final results for each hyperspectral pixel. Not only does the suggested technique benefit from improved extraction of characteristics from CNNs and MS-MA BT, but it also allows complete combined use of visual and temporal data. The performance of the suggested technique is evaluated using SAR data sets, and the results indicate that the MS-MA BT-based multi-functional training algorithm considerably increases identification precision. Recently, convolution neural networks have proved outstanding efficiency on multiple visual activities, including the ranking of common two-dimensional pictures. In this paper, the MS-MA BT multi-scale multi-angle CNN algorithm is used to identify hyperspectral images explicitly in the visual domain. Experimental outcomes based on several SAR image data sets show that the suggested technique can attain greater classification efficiency than some traditional techniques, such as support vector machines and conventional deep learning techniques.

New Estimation Technique for Vehicle-Type-Specific Headway Distributions

1998 ◽  
Vol 1646 (1) ◽  
pp. 18-28 ◽  
Author(s):  
Serge P. Hoogendoorn ◽  
Piet H. L. Bovy
Keyword(s):  
Real Life ◽  
Estimation Procedure ◽  
Distribution Model ◽  
Parameter Estimates ◽  
Passenger Cars ◽  
Car Following ◽  
Vehicle Type ◽  
Real Life Data ◽  
Headway Distribution ◽  
Parameter Values

Recently, a new statistical procedure was developed that enables fast, accurate, and robust estimation of composite headway distributions, such as Branston’s generalized queueing model (GQM). Until now, the new procedure had only been applied to aggregate vehicular flow. In this paper, the estimation procedure is extended to headway observations segregated according to vehicle type and period of the day. Consequently, the parameters of a new mixed-vehicle-type headway distribution model based on Branston’s headway model can be estimated. Distinction of vehicle type and sample periods provides additional insight into the plausibility of the headway distributions and parameter values, as well as into the car-following behavior of the distinct vehicle classes varying across the different periods. The estimation procedure was applied to traffic data collected on a two-lane rural road in the Netherlands. Comparison of the estimated headway distributions with real-life data shows that headway distributions can be realistically replicated with the Pearson-III-based mixed-vehicle-type GQM. Inter-pretable differences between the morning, noon, and evening sample periods and between passenger cars, unarticulated trucks, and articulated trucks are found. In addition, passenger-car equivalents for both articulated trucks and unarticulated trucks were determined from the parameter estimates.

Modeling student knowledge with self-organizing feature maps.

1993 ◽  
Author(s):  
Steven A. Harp ◽  
Tariq Samad ◽  
Michael Villano
Keyword(s):  
Feature Maps ◽  
Student Knowledge ◽  
Self Organizing

scholarly journals A Self-Spatial Adaptive Weighting Based U-Net for Image Segmentation

Electronics ◽  
2021 ◽  
Vol 10 (3) ◽  
pp. 348
Author(s):  
Choongsang Cho ◽  
Young Han Lee ◽  
Jongyoul Park ◽  
Sangkeun Lee
Keyword(s):  
Image Segmentation ◽  
Medical Image ◽  
Medical Image Segmentation ◽  
Feature Maps ◽  
Data Set ◽  
Feature Map ◽  
Adaptive Weighting ◽  
Spatially Adaptive ◽  
Wide Range ◽  
Decoder Architecture

Semantic image segmentation has a wide range of applications. When it comes to medical image segmentation, its accuracy is even more important than those of other areas because the performance gives useful information directly applicable to disease diagnosis, surgical planning, and history monitoring. The state-of-the-art models in medical image segmentation are variants of encoder-decoder architecture, which is called U-Net. To effectively reflect the spatial features in feature maps in encoder-decoder architecture, we propose a spatially adaptive weighting scheme for medical image segmentation. Specifically, the spatial feature is estimated from the feature maps, and the learned weighting parameters are obtained from the computed map, since segmentation results are predicted from the feature map through a convolutional layer. Especially in the proposed networks, the convolutional block for extracting the feature map is replaced with the widely used convolutional frameworks: VGG, ResNet, and Bottleneck Resent structures. In addition, a bilinear up-sampling method replaces the up-convolutional layer to increase the resolution of the feature map. For the performance evaluation of the proposed architecture, we used three data sets covering different medical imaging modalities. Experimental results show that the network with the proposed self-spatial adaptive weighting block based on the ResNet framework gave the highest IoU and DICE scores in the three tasks compared to other methods. In particular, the segmentation network combining the proposed self-spatially adaptive block and ResNet framework recorded the highest 3.01% and 2.89% improvements in IoU and DICE scores, respectively, in the Nerve data set. Therefore, we believe that the proposed scheme can be a useful tool for image segmentation tasks based on the encoder-decoder architecture.

scholarly journals Spectral Convolution Feature-Based SPD Matrix Representation for Signal Detection Using a Deep Neural Network

Entropy ◽  
2020 ◽  
Vol 22 (9) ◽  
pp. 949
Author(s):  
Jiangyi Wang ◽  
Min Liu ◽  
Xinwu Zeng ◽  
Xiaoqiang Hua
Keyword(s):  
Neural Network ◽  
Signal Detection ◽  
Convolutional Neural Network ◽  
Deep Neural Network ◽  
Detection Method ◽  
Learning Algorithm ◽  
Simulated Data ◽  
Data Sets ◽  
Feature Maps ◽  
Simulated Data Sets

Convolutional neural networks have powerful performances in many visual tasks because of their hierarchical structures and powerful feature extraction capabilities. SPD (symmetric positive definition) matrix is paid attention to in visual classification, because it has excellent ability to learn proper statistical representation and distinguish samples with different information. In this paper, a deep neural network signal detection method based on spectral convolution features is proposed. In this method, local features extracted from convolutional neural network are used to construct the SPD matrix, and a deep learning algorithm for the SPD matrix is used to detect target signals. Feature maps extracted by two kinds of convolutional neural network models are applied in this study. Based on this method, signal detection has become a binary classification problem of signals in samples. In order to prove the availability and superiority of this method, simulated and semi-physical simulated data sets are used. The results show that, under low SCR (signal-to-clutter ratio), compared with the spectral signal detection method based on the deep neural network, this method can obtain a gain of 0.5–2 dB on simulated data sets and semi-physical simulated data sets.

Sign in / Sign up

Export Citation Format

Share Document