Enhancing the universality and transferability of freeway incident detection using a Bayesian-based neural network

Abdulhai, Baher; Ritchie, Stephen G.

doi:10.1016/s0968-090x(99)00022-4

Cited by 106 publications

(58 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The other most well-known ANN algorithm used for detecting traffic incidents is the Probabilistic Neural Network (PNN) structures (Abdulhai and Ritchie, 1999a;Abdulhai and Ritchie, 1999b;Jin et al, 2002). The PNN structure consists of four layers, the input layer, pattern layer, summation layer, and the output layer.…”

Section: A Multi-layer Feed-forward Neural Networkmentioning

confidence: 99%

“…In more recent years Artificial Intelligence (AI) based incident detection algorithms have been developed to tackle the problem of incident detection. The most well-known AI based algorithms are based on Artificial Neural Networks (ANN) (Cheu et al, 1991;Ritchie and Cheu, 1993;Cheu and Ritchie, 1995;Stephanedes and Liu, 1995;Dia and Rose, 1997;Abdulhai and Ritchie, 1999a;Abdulhai and Ritchie, 1999b;Jin et al, 2002;Srinivasan et al 2008), fuzzy logic (Chang and Wang, 1994;Lin and Chang, 1998) and support vector machine (SVM) algorithms (Yuan and Cheu, 2003). 3 However, the majority of the research attention in incident detection algorithms has been generally focussed towards freeway incident management.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Customization of Automatic Incident Detection Algorithms for Signalized Urban Arterials

Ghosh

Smith

2014

Journal of Intelligent Transportation Systems

View full text Add to dashboard Cite

Non-recurrent congestion or incidents are detrimental to the operability and efficiency of busy urban transport networks. There exists multiple Automatic Incident Detection Algorithms (AIDA) to remotely detect the occurrence of an incident in highway or freeway scenarios, however very little research has been performed to automatically detect incidents in signalised urban arterials. This limited research attention has mostly been focussed on developing new urban arterial specific algorithms rather than identifying alternative methods to synthesize existing freeway based algorithms to urban conditions. The main hindrance to such synthesis is that the traffic patterns on the signalised urban arterials are significantly different from the same on highways/freeways due to the presence of traffic intersections. This paper introduces a new strategy of customising the existing AIDAs (freeway based or otherwise) to significantly improve their adaptability to signalised urban arterial transport networks. The new strategy focuses on preprocessing the traffic information before being used as input to a freeway/highway based AIDA to lessen the effect of traffic signals and to imitate the input patterns in highway/freeway based incident conditions. The effectiveness of this new strategy has been established with the help of four existing AIDAs. The proposed strategy is a simple solution to implement existing algorithms to signalised urban networks without any further instrumentation or operational cost.

show abstract

Section: A Multi-layer Feed-forward Neural Networkmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Customization of Automatic Incident Detection Algorithms for Signalized Urban Arterials

Ghosh

Smith

2014

Journal of Intelligent Transportation Systems

View full text Add to dashboard Cite

show abstract

“…In [5], the authors used genetic algorithm to optimize the parameter search for probabilistic neural network (PNN) and the result showed that this approach achieved better DR and FAR. A modified form of the Bayesian-based PNN was proposed in [6] and the study result indicated that this method enhances the universality and transferability of the detection algorithm.…”

Section: An Unsupervised Feature Learning Approach To Improve Automatmentioning

confidence: 99%

An unsupervised feature learning approach to improve automatic incident detection

Ren

Wang

Wang³

et al. 2012

2012 15th International IEEE Conference on Intelligent Transportation Systems

View full text Add to dashboard Cite

Abstract-sophisticated automatic incident detection (AID) technology plays a key role in contemporary transportation systems. Though many papers were devoted to study incident classification algorithms, few study investigated how to enhance feature representation of incidents to improve AID performance. In this paper, we propose to use an unsupervised feature learning algorithm to generate higher level features to represent incidents. We used real incident data in the experiments and found that effective feature mapping function can be learnt from the data crosses the test sites. With the enhanced features, detection rate (DR), false alarm rate (FAR) and mean time to detect (MTTD) are significantly improved in all of the three representative cases. This approach also provides an alternative way to reduce the amount of labeled data, which is expensive to obtain, required in training better incident classifiers since the feature learning is unsupervised.

show abstract

“…In the Probabilistic Neural Network (PNN) algorithm (Abdulhai & Ritchie, 1999a;Abdulhai & Ritchie, 1999b;Jin, Cheu, & Srinivasan, 2002), the transfer function of the hidden layer is a radial-basis function, and that for the output layer is a competitive-transfer function. The PNN consists of four layers -the input layer, the pattern layer, the summation layer, and the output layer.…”

Section: Probabilistic Neural Network Algorithmmentioning

confidence: 99%

The Role of Non-Recurring Congestion in Massive Hurricane Evacuation Events

Fonseca¹,

Moynihan²,

Fernandes³

2011

Recent Hurricane Research - Climate, Dynamics, and Societal Impacts

View full text Add to dashboard Cite

Enhancing the universality and transferability of freeway incident detection using a Bayesian-based neural network

Cited by 106 publications

References 4 publications

Customization of Automatic Incident Detection Algorithms for Signalized Urban Arterials

Customization of Automatic Incident Detection Algorithms for Signalized Urban Arterials

An unsupervised feature learning approach to improve automatic incident detection

The Role of Non-Recurring Congestion in Massive Hurricane Evacuation Events

Contact Info

Product

Resources

About