Traffic signs recognition based on visual attention mechanism

Guo, Hairu; Wang, Xiaojie; Zhong, Yiwen; Peng, Lü

doi:10.1016/s1005-8885(10)60139-2

Cited by 13 publications

(8 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This opens up an interesting topic concerning the influence of the shape of a specific symbol on an eHMI on it's visual attention. Here, traffic signs can be used as a comparison, which stand out from their surroundings due to their salient shape (triangular, round, angular) or color and can therefore be recognized quickly [45]. Accordingly, a possible salient symbol form for eHMIs should be in the focus of further investigations.…”

Section: B Decision Times Of Subjectsmentioning

confidence: 99%

Displaying the Driving State of Automated Vehicles to Other Road Users: An International, Virtual Reality-Based Study as a First Step for the Harmonized Regulations of Novel Signaling Devices

Singer

Kobbert

Zandi

et al. 2022

IEEE Trans. Intell. Transport. Syst.

View full text Add to dashboard Cite

Additional signaling devices for highly automated vehicles (AVs) that can communicate their driving state to other road users can simplify the integration process in existing road traffic. This paper presents the results of an international, virtual reality-based study conducted in China, South Korea and the USA in which subjects assume the role of a pedestrian and are placed in direct encounter situations with an AV in a parking lot. A novel communication interface consisting of three displays is attached to the AV's front and used to show additional information about its driving state. In total, three encounter scenarios are investigated: the AV approaches from the left, front and right outside of the pedestrian's line of sight. The influence of different symbol types on the subject's moving behavior, recognition of intention and perceived safety is investigated. The results show that additional signals ensure a better perception of the AV's intention and increase the perceived safety. The moving behavior of subjects is significantly changed when additional signals are used during driving tasks compared to the same tasks without such signals. The change of moving behaviour is similar in encounter situations where the AV approaches from the left and front but differs in encounter situations from the right. These results could equally be proven for all nationalities, which shows that a uniform, international solution for additional signaling devices of highly automated vehicles is possible.

show abstract

Section: B Decision Times Of Subjectsmentioning

confidence: 99%

Displaying the Driving State of Automated Vehicles to Other Road Users: An International, Virtual Reality-Based Study as a First Step for the Harmonized Regulations of Novel Signaling Devices

Singer

Kobbert

Zandi

et al. 2022

IEEE Trans. Intell. Transport. Syst.

View full text Add to dashboard Cite

show abstract

“…With the proposal of attention mechanism [ 37 ] and the rapid development in the field of image, attention mechanism is gradually combined with natural language processing. Especially in machine translation, attention mechanism is introduced between the current state of the target language sequence and the hidden layer state of the source language sequence.…”

Section: Related Workmentioning

confidence: 99%

Short Sequence Chinese-English Machine Translation Based on Generative Adversarial Networks of Emotion

Wang

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

With the steady growth of the global economy, the communication between countries in the world has become increasingly close. Due to its translation efficiency and other problems, the traditional manual translation has gradually failed to meet the current people’s translation requirements. With the rapid development of machine-learning and deep-learning related technologies, artificial intelligence-related technologies have affected various industries, including the field of machine translation. Compared with traditional methods, neural network-based machine translation has high efficiency, so this field has attracted many scholars’ intensive research. How to improve the accuracy of neural machine translation through deep learning technology is the core problem that researchers study. In this paper, the neural machine translation model based on generative adversarial network is studied to make the translation result of neural network more accurate and three-dimensional. The model uses adversarial thinking to consider the sequence of emotion direction so that the translation results are more humanized. We set up several experiments to verify the efficiency of the model, and the experimental results prove that the proposed model is suitable for Chinese-English machine translation.

show abstract

“…Graph Convolutional Networks (GCN) [7,8] is used to extract the multipoint series space spatial features, which increases the expression ability of the model spatial dimension, and uses IEALL model to extract the change trend of the temporal dimension, thereby increasing the prediction accuracy of the model. STAGCN-IEALL also introduces a spatio-temporal attention mechanism [9][10][11][12] to capture the dynamic correlation from both spatial and temporal dimensions.…”

Section: Introductionmentioning

confidence: 99%

IEALL: Dam Deformation Prediction Model Based on Combination Model Method

Jing

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

The accuracy of dam deformation prediction is a key issue that needs to be addressed due to the many factors that influence dam deformation. In this paper, a dam deformation prediction model based on IEALL (IGWO-EEMD-ARIMA-LSTM-LSTM) is proposed for a single-point scenario. The IEALL model is based on the idea of a combination model. Firstly, EEMD is used to decompose the dam deformation data, and then the ARIMA and LSTM models are selected for prediction. To address the problem of low prediction accuracy caused by simple linear addition of prediction results from different models in traditional combination models, the LSTM model is used to learn the combination relationship of different model prediction results. The problem of neural network parameters falling into local optima due to random initialization is addressed by using the improved gray wolf optimization (IGWO) to optimize multiple parameters in the IEALL combination model to obtain the optimal parameters. For the multi-point scenario of dam deformation, based on the IEALL model, a dam deformation prediction model based on spatio-temporal correlation and IEALL (STAGCN-IEALL) is proposed. This model introduces graph convolutional neural networks (GCN) to extract spatial features from multi-point sequences, increasing the model’s ability to express spatial dimensions. To address the dynamic correlation between different points in the deformation sequence at any time and the dynamic dependence on different points at any given time, spatio-temporal attention mechanisms are introduced to capture dynamic correlation from both spatial and temporal dimensions. Experimental results showed that compared to ST-GCN, IEALL reduced the RMSE, MAE, and MAPE by 16.06%, 14.72%, and 21.19%. Therefore, the proposed model effectively reduces the prediction error and can more accurately predict the trend of dam deformation changes.

show abstract

Traffic signs recognition based on visual attention mechanism

Cited by 13 publications

References 6 publications

Displaying the Driving State of Automated Vehicles to Other Road Users: An International, Virtual Reality-Based Study as a First Step for the Harmonized Regulations of Novel Signaling Devices

Displaying the Driving State of Automated Vehicles to Other Road Users: An International, Virtual Reality-Based Study as a First Step for the Harmonized Regulations of Novel Signaling Devices

Short Sequence Chinese-English Machine Translation Based on Generative Adversarial Networks of Emotion

IEALL: Dam Deformation Prediction Model Based on Combination Model Method

Contact Info

Product

Resources

About