Real-Time Vision-Based Vehicle Detection and Tracking on a Moving Vehicle for Nighttime Driver Assistance

Chen, Yen‐Lin; Wu, Bo-Xun; Lin, Chia-How; Fan, Chung-Jui; Hsieh, Chaur-Heh

doi:10.2316/journal.206.2009.2.206-3089

Cited by 4 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The first step aimed to search the ROI of vehicles and the second step was to determine their effective ROI. With regard to the first step, multilevel thresholding approaches ( 3 , 19 , 20 ), paired vehicle lights ( 21 , 22 ), salient maps ( 6 ), and AdaBoost classifier were included ( 5 ). As for determining the effective ROI of vehicles, Hough transform ( 2 , 23 ), night DPM ( 24 ), and Laplacian of Gaussian operator ( 4 ) and so forth were developed.…”

Section: Related Workmentioning

confidence: 99%

Improved Nighttime Traffic Detection Using Day-To-Night Image Transfer

Guo

Jian

Xie

et al. 2023

Transportation Research Record: Journal of the Transportation R

View full text Add to dashboard Cite

With the rapid development of convolutional neural networks (CNNs), real-time traffic monitoring, which benefits both traffic optimization and management, has been widely applied with smart cameras on streets. However, compared with the daytime traffic detection, nighttime traffic detection is limited by labeling annotations and is thus unstable, inaccurate, and inefficient in practice. To address this issue, this study proposes to use the augmented nighttime traffic images generated by cycle generative adversarial networks (CycleGAN) for better nighttime detection performance. Using CycleGAN, transferred nighttime traffic images are generated by using the daytime traffic images, as both share the same annotations of traffic instances in the daytime. For comparison, different learning rates and crop sizes are adopted for day-to-night traffic image transfer. The previously proposed detection network, dense traffic detection network (DTDNet), is adopted to train the prepared image data set. The indicators of mean average precision (mAP), precision, and recall are adopted for training performance evaluation and comparison. Based on the visualization results, CycleGAN, with a learning rate of 2e-5 and a crop size of 64, has better performance on day-to-night traffic image transfer with our proposed image data set. Considering the indicators of training performance, DTDNet, with 60% of transferred nighttime images and 40% of the original images, has better accuracy in four categories. Overall, this study provides a possible solution for addressing the issue of training data limitation in nighttime traffic detection and demonstrates the potential of GAN-based data augmentation in the transportation domain.

show abstract

Section: Related Workmentioning

confidence: 99%

Improved Nighttime Traffic Detection Using Day-To-Night Image Transfer

Guo

Jian

Xie

et al. 2023

Transportation Research Record: Journal of the Transportation R

View full text Add to dashboard Cite

show abstract

“…To resolve this issue gaussian mixture modelling (GMM) is used in combination with expectation optimization so that there is developing the background model [4]. The occultation was also developed by using colored histograms so that it helps to propose and identify the number of vehicles in dark videos by removing headlights [5]. The offline training method also helps to detect the vehicle by using a support vector machine so that it estimates the Haar wavelet function [6].…”

Section: Introductionmentioning

confidence: 99%

Application of improved you only look once model in road traffic monitoring system

Kamble

Kounte

2023

IJECE

View full text Add to dashboard Cite

<span lang="EN-US">The present research focuses on developing an intelligent traffic management solution for tracking the vehicles on roads. Our proposed work focuses on a much better you only look once (YOLOv4) traffic monitoring system that uses the CSPDarknet53 architecture as its foundation. Deep-sort learning methodology for vehicle multi-target detection from traffic video is also part of our research study. We have included features like the Kalman filter, which estimates unknown objects and can track moving targets. Hungarian techniques identify the correct frame for the object. We are using enhanced object detection network design and new data augmentation techniques with YOLOv4, which ultimately aids in traffic monitoring. Until recently, object identification models could either perform quickly or draw conclusions quickly. This was a big improvement, as YOLOv4 has an astoundingly good performance for a very high frames per second (FPS). The current study is focused on developing an intelligent video surveillance-based vehicle tracking system that tracks the vehicles using a neural network, image-based tracking, and YOLOv4. Real video sequences of road traffic are used to test the effectiveness of the method that has been suggested in the research. Through simulations, it is demonstrated that the suggested technique significantly increases graphics processing unit (GPU) speed and FSP as compared to baseline algorithms.</span>

show abstract

“…Also in [11] the distance between two tail lights as a source of information to the detection of cars in night time is considered; these detections do not take place in real time but through videos recorded previously. In the same way in [12] the tail light is segmented through a multilevel threshold histogram as a previous stage to the detection of vehicles and the estimation of distances. Following the same line in [13] a series of sensors to assist in the night time driving are implemented.…”

Section: Related Workmentioning

confidence: 99%

Night Time and Low Visibility Driving Assistance Based on the Application of Colour and Geometrical Features Extraction

et al. 2016

View full text Add to dashboard Cite

Abstract. The present work shows an application to detect cars in night environments as a means to assist the car driving through HSV (Hue, Saturation, Value) colour extraction and geometric modelling. The developed algorithm has been implemented in smart devices through the platform Android. The detection and tracking of vehicles are implemented in low visibility environments such as night time, raining or snowing conditions; the different tests carried out confirm high performance rates of detection (p = 95.2 %). The information provided by the different sensors of the smart devices have been used to generate virtual information in a real driving environment (Augmented Reality) in order to complement the functionalities of the purposed solution. This information consists of visual, vibratory and auditory warnings that detect possible collisions and dangerous driving situations.

show abstract

Real-Time Vision-Based Vehicle Detection and Tracking on a Moving Vehicle for Nighttime Driver Assistance

Cited by 4 publications

References 0 publications

Improved Nighttime Traffic Detection Using Day-To-Night Image Transfer

Improved Nighttime Traffic Detection Using Day-To-Night Image Transfer

Application of improved you only look once model in road traffic monitoring system

Night Time and Low Visibility Driving Assistance Based on the Application of Colour and Geometrical Features Extraction

Contact Info

Product

Resources

About