A Study on Real-Time Detection Method of Lane and Vehicle for Lane Change Assistant System Using Vision System on Highway

Nguyen, Van-Quang; Kim, Heungsuk; Jun, SeoChang; Boo, Kwangsuck

doi:10.1016/j.jestch.2018.06.006

Cited by 45 publications

(21 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this context, as of yet unanswered research questions on design guidelines and the acceptance of AI-based digital assistance is posed, while the orientation on the maxim of "ethics-by-design" is demanded [22]. While some research exists on utilizing computer vision approaches in assistance systems [32,33,34], there is a need to further investigate the conjunction of computer vision, assistance systems and AI applications with human intelligence in the loop. This paper will therefore provide further evidence on the efficiency and effectiveness of such human-AI collaboration systems in a computer vision use case.…”

Section: Human-ai Collaborationmentioning

confidence: 99%

Deep learning object detection as an assistance system for complex image labeling tasks

Leimkühler

Gravemeier

Biester³

et al. 2021

Proceedings of the Annual Hawaii International Conference on System Sciences

View full text Add to dashboard Cite

Object detection via deep learning has many promising areas of application. However, robustness and accuracy of fully automated systems are often insufficient for practical use. Integrating results from Artificial Intelligence (AI) and human intelligence in collaborative settings might bridge the gap between efficiency and accuracy. This study proves increased efficiency when supporting human intelligence through AI without negative impact on effectiveness in a finegrained car scratch image labeling task. Based on the confirmed benefits of AI with human intelligence in the loop approaches, this contribution discusses potential practical application scenarios and envisions the implementation of assistance systems supported by computer vision.

show abstract

Section: Human-ai Collaborationmentioning

confidence: 99%

Deep learning object detection as an assistance system for complex image labeling tasks

Leimkühler

Gravemeier

Biester³

et al. 2021

Proceedings of the Annual Hawaii International Conference on System Sciences

View full text Add to dashboard Cite

show abstract

“…This size difference is due to the fact that when sampling a small portion of the image it is often difficult to find out what is shown on it. To eliminate the problems of retraining, the DropOutmethod is included in the fifth fully connected network layer [18]. This method consists in randomly allocating some area of the neural network in which weights are updated.…”

Section: Neural Network Trainingmentioning

confidence: 99%

“…We also used the L2 network regularization method, which consists in a large fine of too high a weight value and a small one at a low value. Also, during training, the loss function was minimized using the Minibatch gradient descent method [18].…”

Section: Neural Network Trainingmentioning

confidence: 99%

“…Studyingtime Accuracy (%) h. min. Nesterovacceleratedgradient [16] 14 46 85,56 AdaGrad [17] 30 82,24 Adam [18] 28 89,2 The results presented in table 1 show that the Adam algorithm showed the best results relative to the rest: training time 14 hours 28 minutes, classification accuracy 89,2%. Accuracy for each class of objects: roads 90,89%, buildings 94,65%, trees 89,89%, bushes and grass 80,62%, cars 89,32%.…”

Section: Algorithmnamementioning

confidence: 99%

See 1 more Smart Citation

Semantic segmentation algorithms of the Earth's surface pictures based on neural network methods

Druki¹,

Spitsyn²,

Аркалыков³

2020

Tomsk State University Journal of Control and Computer Science

View full text Add to dashboard Cite

show abstract

“…New emerging techniques based on Deep Learning such as Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Long short-term memory (LSTM) networks have also been proposed [14,20,21,22,23]. The problem of vehicle detection and counting has been addressed from different points of view, such as top view [12,21], top-front view [15,16] and even on-road view for autonomous cars [18,24].…”

Section: Introductionmentioning

confidence: 99%

Vehicle Counting in Video Sequences: An Incremental Subspace Learning Approach

Rosas-Arias

Portillo-Portillo

Olivares-Mercado

et al. 2019

Sensors

View full text Add to dashboard Cite

The counting of vehicles plays an important role in measuring the behavior patterns of traffic flow in cities, as streets and avenues can get crowded easily. To address this problem, some Intelligent Transport Systems (ITSs) have been implemented in order to count vehicles with already established video surveillance infrastructure. With this in mind, in this paper, we present an on-line learning methodology for counting vehicles in video sequences based on Incremental Principal Component Analysis (Incremental PCA). This incremental learning method allows us to identify the maximum variability (i.e., motion detection) between a previous block of frames and the actual one by using only the first projected eigenvector. Once the projected image is obtained, we apply dynamic thresholding to perform image binarization. Then, a series of post-processing steps are applied to enhance the binary image containing the objects in motion. Finally, we count the number of vehicles by implementing a virtual detection line in each of the road lanes. These lines determine the instants where the vehicles pass completely through them. Results show that our proposed methodology is able to count vehicles with 96.6% accuracy at 26 frames per second on average—dealing with both camera jitter and sudden illumination changes caused by the environment and the camera auto exposure.

show abstract

A Study on Real-Time Detection Method of Lane and Vehicle for Lane Change Assistant System Using Vision System on Highway

Cited by 45 publications

References 11 publications

Deep learning object detection as an assistance system for complex image labeling tasks

Deep learning object detection as an assistance system for complex image labeling tasks

Semantic segmentation algorithms of the Earth's surface pictures based on neural network methods

Vehicle Counting in Video Sequences: An Incremental Subspace Learning Approach

Contact Info

Product

Resources

About