Object Detection-Based License Plate Localization and Recognition in Complex Environments

Tao, Ting; Dong, Dezun; Huang, Shize; Chen, Wei; Yang, Liansheng

doi:10.1177/0361198120954202

Cited by 10 publications

(3 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similarly, in 2019, in the United States, Aliari and Sadabadi (13) applied the faster CNN (FCNN) method to analyze speed data over time and to identify congestion on highways. In 2020, in China, Tao et al (14) applied the YOLOv3 technique to recognize license plates and found that it recognized more than 99% of letters and numbers.…”

Section: Vehicle Passenger Detection Technologymentioning

confidence: 99%

Enhancing the Performance of Vehicle Passenger Detection under Adverse Weather Conditions Using Augmented Reality-Based Machine Learning Approach

Lee¹,

Kang²,

Lim³

et al. 2021

Transportation Research Record: Journal of the Transportation R

View full text Add to dashboard Cite

In response to extreme traffic congestion in metropolitan areas that causes unnecessarily long travel times, high fuel consumption, and excessive greenhouse gas emissions, transportation agencies have implemented various strategies to mitigate traffic congestion. Managed lanes—one of the measures applied worldwide—provide benefits to road users and operating agencies by integrating advanced technologies such as electronic and dynamic tolling systems. However, those agencies already implementing or considering implementing the managed lane strategy are seeking a solution to effectively and properly charge toll rates based on vehicle occupancy and penalize violating vehicles. Vehicle passenger detection systems (VPDSs) have been developed and evaluated worldwide, but limitations still inhibit their full implementation. This study confirms that the performance of the deep learning algorithm, a core VPDS technology, declines under certain adverse weather conditions because of lack of training data sets. The performance of the “you only look once” (YOLOv3) model trained with a normal weather data set decreased by as much as 8.5% when it was tested for adverse weather conditions. In this study, augmented reality (AR) models are developed to enhance the accuracy of vehicle passenger detection (VPDA) by the VPDS by training the algorithm with AR images representing virtual adverse weather conditions. Models trained with AR image sets of various weather categories (fog, rain, and snow) attained VPDA enhanced by up to 7.9%. The final model significantly improves VPDA under adverse weather conditions. The proposed models could be considered for implementation with road weather information systems under adverse weather conditions.

show abstract

Section: Vehicle Passenger Detection Technologymentioning

confidence: 99%

Enhancing the Performance of Vehicle Passenger Detection under Adverse Weather Conditions Using Augmented Reality-Based Machine Learning Approach

Lee¹,

Kang²,

Lim³

et al. 2021

Transportation Research Record: Journal of the Transportation R

View full text Add to dashboard Cite

show abstract

“…It is becoming increasingly common to use convolutional neural networks (CNN) in image recognition, such as image classification 1,2 and object detection. 3 Human pose estimation (HPE) 4 is a CNN-based computer vision technique, which uses a blend of regression and classification approaches to detect the composition of human bodies. It is possible, for instance, to adopt computer vision techniques based on HPE to help supervisors identify the behavior of humans.…”

Section: Introductionmentioning

confidence: 99%

Adversarial attack on human pose estimation network

Zhang,

Huang,

Liu

et al. 2024

J. Electron. Imag.

Self Cite

View full text Add to dashboard Cite

Real-time human pose estimation (HPE) using convolutional neural networks (CNN) is critical for enabling machines to better understand human beings based on images and videos, and for assisting supervisors in identifying human behavior. However, CNN-based systems are susceptible to adversarial attacks, and the attacks specifically targeting HPE have received little attention. We present a gradient-based adversarial example generation method, named AdaptiveFool, which is designed to effectively perform a keypoints-invisible attack against OpenPose by aggregating the loss function of human keypoints and generating adaptive adversarial perturbations. In addition, we introduce an object-oriented perturbation generation method during the AdaptiveFool process to eliminate background perturbations. Our proposed method adapts the adversarial perturbations and generates object-oriented perturbations. On COCO 2017 datasets, our method achieves 6.3% mean average precision on OpenPose. This research provides inspiration for future work on developing efficient and effective adversarial example defense methods for HPE.

show abstract

“…In the current intelligent traffic field, accurate recognition of license plate information is not only conducive to the handling of traffic accidents, but also beneficial to traffic safety [1][2][3][4]. However, in the actual traffic situation, due to the installation position and established angle of the camera, the license plate in the camera image has a certain tilt angle in both horizontal and vertical directions [5][6][7]. This brings a lot of difficulties to license plate information recognition.…”

Section: Introductionmentioning

confidence: 99%

Tilt Angle Detection and Correction of License Plate Based on Convolutional Neural Network

2022

IJOMAM

View full text Add to dashboard Cite

In the intelligent traffic field, accurate recognition of license plate information is not only conducive to the handling of traffic accidents, but also beneficial to safety in the autonomous field. However, the identification of license plate information becomes very difficult if there are tilted angles in the license plate image. The objective of the work is to establish a convolutional neural network (CNN)-based license plate tilt angle recognition network realizing the horizontal and vertical tilt angles detection of the license plate accurately; and then, according to the detected angles, the tilt angle detection algorithm is proposed to realize the tilted license plate image correction and character segmentation. Firstly, the YOLOv3 is adopted to locate license plate and classify the clockwise rotation and anticlockwise rotation license plate images. Secondly, the regression CNN is established to detect the horizontal and vertical tilt angle of the cropped license plate image. Thirdly, the tilted license plate correction algorithm is proposed to correct the horizontal and vertical tilt angles of the license plate using image rotation and shearing transformation theory according to the detected horizontal and vertical tilt angles. Subsequently, the bilinear interpolation is used to improve the correction performance. Finally, the characters of the license plate are segmented from the corrected license plate image after removing the upper and lower boundary. The results show that the proposed tilt angle detection network and the correction algorithm have a good performance, which is superior to the current state-of-the-art tilted license plate correction algorithm.

show abstract

Object Detection-Based License Plate Localization and Recognition in Complex Environments

Cited by 10 publications

References 18 publications

Enhancing the Performance of Vehicle Passenger Detection under Adverse Weather Conditions Using Augmented Reality-Based Machine Learning Approach

Enhancing the Performance of Vehicle Passenger Detection under Adverse Weather Conditions Using Augmented Reality-Based Machine Learning Approach

Adversarial attack on human pose estimation network

Tilt Angle Detection and Correction of License Plate Based on Convolutional Neural Network

Contact Info

Product

Resources

About