Traffic Congestion Classification for Nighttime Surveillance Videos

Chen, Hua-Tsung; Tsai, Li-Wu; Gu, Hui-Zhen; Lee, Suh-Yin; Lin, Bor‐Shyh

doi:10.1109/icmew.2012.36

Cited by 9 publications

(4 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The vehicle flow feature always refers to the count of vehicles in a time period. Reference [21] proposed a method to count entering cars using virtual line, the traffic condition can then be identified based on the number of vehicles. In [22], a method based on background modeling and vehicle detection was proposed.…”

Section: A Methods Using Successive Video Framesmentioning

confidence: 99%

Exploiting Deeply Supervised Inception Networks for Automatically Detecting Traffic Congestion on Freeway in China Using Ultra-Low Frame Rate Videos

Sun

Wang

Wang³

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Traffic congestion detection plays an important role for road management. However, it is difficult to automatically report traffic congestion when it occurs in large-scale road network. One of key challenges for rapidly and precisely identifying early congestion is huge variations in appearance caused by illumination, weather, camera settings and other traffic conditions. To address it, we proposed a trafficoriented model to classify congestion from large dataset of ultra-low frame rate video captured from traffic surveillance system. The proposed deeply supervised traffic congestion detector has two modules: attention proposal module and deeply supervised inception network. Specifically, within the shallow layers, the binary edge/corner density features are used in attention proposal module to generate the rang of interest (ROI) mask automatically. This strategy keeps the training process focusing on the congestion features without disturbances. Following the attention proposal module, a very deep structure based on the inception network was used together to effectively extract rich and discriminative features then detect traffic congestion. The approach was tested on a self-established dataset based on empirical data, which contains images captured from 14470 surveillance cameras for monitoring 5,215 km of freeway in Shaanxi province, China. The experimental results show that the accuracy of the proposed method could reach 95.77% considering various disturbances, conditions and other limitations, which is improved than unsupervised networks.

show abstract

Section: A Methods Using Successive Video Framesmentioning

confidence: 99%

Exploiting Deeply Supervised Inception Networks for Automatically Detecting Traffic Congestion on Freeway in China Using Ultra-Low Frame Rate Videos

Sun

Wang

Wang³

et al. 2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Our code (in Matlab, Python & C/C++) is provided on our GitHub repository [44]. 1 Learning achieved by taking the convolutional base of a pre-trained network, running the new data of 4 traffic categories through it and training a new randomly initialized classifier 2 Learning achieved by re-training the classifier with wrong predictions till operating period of the system 3 Even though GPUs could be an efficient accelerators for CNNs. However, such devices are expensive & very power hungry and thus, make them not suitable in the aforementioned power-constrained scenarios 4 RTL: Register-transfer level is a design abstraction, which models a synchronous digital circuit in terms of the flow of digital data between hardware registers and the logical operations performed on those signals.…”

Section: A Hardware Infrastructure and Software Infrastructurementioning

confidence: 99%

MAT-CNN-SOPC: Motionless Analysis of Traffic Using Convolutional Neural Networks on System-On-a-Programmable-Chip

Dey

Kalliatakis

Saha

et al. 2018

2018 NASA/ESA Conference on Adaptive Hardware and Systems (AHS)

View full text Add to dashboard Cite

Intelligent Transportation Systems (ITS) have become an important pillar in modern "smart city" framework which demands intelligent involvement of machines. Traffic load recognition can be categorized as an important and challenging issue for such systems. Recently, Convolutional Neural Network (CNN) models have drawn considerable amount of interest in many areas such as weather classification, human rights violation detection through images, due to its accurate prediction capabilities. This work tackles real-life traffic load recognition problem on System-On-a-Programmable-Chip (SOPC) platform and coin it as MAT-CNN-SOPC, which uses an intelligent retraining mechanism of the CNN with known environments. The proposed methodology is capable of enhancing the efficacy of the approach by 2.44x in comparison to the state-of-art and proven through experimental analysis. We have also introduced a mathematical equation, which is capable of quantifying the suitability of using different CNN models over the other for a particular application based implementation.Index Terms-Convolutional neural network (CNN), traffic analysis, traffic density, transfer learning, system-on-aprogrammable-chip (SOPC).

show abstract

“…Traffic conditions can be monitored by the cameras installed along roads and highways to provide realtime traffic information, such as average driving speed and traffic flow, assisting drivers in avoiding congestion areas [5], [6]. Furthermore, traffic accidents or violations can also be detected, informing the police promptly [7].…”

Section: Introductionmentioning

confidence: 99%

Multi-camera vehicle identification in tunnel surveillance system

Chen

Chu

Chou

et al. 2015

2015 IEEE International Conference on Multimedia &Amp; Expo Workshops (ICMEW)

Self Cite

View full text Add to dashboard Cite

Tunnel traffic security has received increasing attention since accidents in tunnels may cause serious casualties. Surveillance cameras are widely equipped in tunnels for traffic condition monitoring and safety maintenance. Vehicle identification among multiple cameras is an essential component in tunnel surveillance systems. In this paper, we propose a Spatiotemporal Successive Dynamic Programming (S2DP) algorithm for identifying vehicles between pairs of cameras. Taking color information into consideration, we extract features based on Harris corner detection with OpponentSIFT descriptors. "Tracking-by-identification " for vehicles across multiple cameras can thus be achieved. Extensive experiments on real tunnel video data show that the proposed S2DP algorithm outperforms state-of-the-art methods.

show abstract

Traffic Congestion Classification for Nighttime Surveillance Videos

Cited by 9 publications

References 8 publications

Exploiting Deeply Supervised Inception Networks for Automatically Detecting Traffic Congestion on Freeway in China Using Ultra-Low Frame Rate Videos

Exploiting Deeply Supervised Inception Networks for Automatically Detecting Traffic Congestion on Freeway in China Using Ultra-Low Frame Rate Videos

MAT-CNN-SOPC: Motionless Analysis of Traffic Using Convolutional Neural Networks on System-On-a-Programmable-Chip

Multi-camera vehicle identification in tunnel surveillance system

Contact Info

Product

Resources

About