Lightweight fully convolutional network for license plate detection

Han, Xiang; Zhao, Yong; Yuan, Yule; Zhang, Guiying; Hu, Xuefeng

doi:10.1016/j.ijleo.2018.10.098

Cited by 23 publications

(16 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We used 80 images of the Caltech Cars dataset for training and 46 for testing, as in [55][56][57]. Then, we employed 16 of the 80 training images for validation (i.e., 20%).…”

Section: Evaluation Protocolmentioning

confidence: 99%

A Robust Real-Time Automatic License Plate Recognition Based on the YOLO Detector

Laroca

Severo

Zanlorensi

et al. 2018

2018 International Joint Conference on Neural Networks (IJCNN)

413

271

View full text Add to dashboard Cite

In this paper, we present an efficient and layout-independent Automatic License Plate Recognition (ALPR) system based on the stateof-the-art YOLO object detector that contains a unified approach for license plate (LP) detection and layout classification to improve the recognition results using post-processing rules. The system is conceived by evaluating and optimizing different models with various modifications, aiming at achieving the best speed/accuracy trade-off at each stage. The networks are trained using images from several datasets, with the addition of various data augmentation techniques, so that they are robust under different conditions. The proposed system achieved an average end-to-end recognition rate of 96.8% across eight public datasets (from five different regions) used in the experiments, outperforming both previous works and commercial systems in the ChineseLP, OpenALPR-EU, SSIG-SegPlate and UFPR-ALPR datasets. In the other datasets, the proposed approach achieved competitive results to those attained by the baselines. Our system also achieved impressive frames per second (FPS) rates on a high-end GPU, being able to perform in real time even when there are four vehicles in the scene. An additional contribution is that we manually labeled 38,351 bounding boxes on 6,239 images from public datasets and made the annotations publicly available to the research community.

show abstract

“…We used 80 images of the Caltech Cars dataset for training and 46 for testing, as in [55][56][57]. Then, we employed 16 of the 80 training images for validation (i.e., 20%).…”

Section: Evaluation Protocolmentioning

confidence: 99%

A Robust Real-Time Automatic License Plate Recognition Based on the YOLO Detector

Laroca

Severo

Zanlorensi

et al. 2018

2018 International Joint Conference on Neural Networks (IJCNN)

413

271

View full text Add to dashboard Cite

show abstract

“…By analyzing the above algorithms, the accuracy of the direct positioning algorithm is lower than that of the indirect positioning algorithm for most time. For example, the accuracy of [53] is 87% and 93.47% of [56] are both lower than that of the indirect positioning algorithm [39], [59], [63], [64]. In table 3, we summarize above existing detection algorithms and analysis the advantages and disadvantages of each algorithm.…”

Section: Discussionmentioning

confidence: 99%

“…Xiang et al [56] presented an efficient lightweight full convolutional network for license plate detection from complex scenes, which downscales input images for substantially accelerating proceeding and reducing the computational cost. In order to further improve the prediction accuracy, dense connections and dilated convolutions are adopted for combing multi-level and multi-scale vision features, and the fusion loss structure is appended during training.…”

Section: A Direct Locationmentioning

confidence: 99%

Research on License Plate Recognition Algorithms Based on Deep Learning in Complex Environment

Wang

2020

IEEE Access

View full text Add to dashboard Cite

License plate recognition systems are widely used in modern smart cities, such as toll payment systems, parking fee payment systems and residential access control. Such electronic systems are not only convenient for people's daily life, but also provide safe and efficient services for managers. License plate recognition algorithm is a mature but imperfect technology. The traditional location recognition algorithm is easily affected by light, shadow, background complexity or other factors, resulting in the failure to meet the application of real scenes. With the development of deep learning, the license plate recognition algorithm can extract deeper features, thus greatly improving the detection and recognition accuracy. Therefore, this paper discusses the application of deep learning in license plate recognition, and the main work is as follows: 1) Introduce the most advanced algorithms from the three main technical difficulties: license plate skew, image noise and license plate blur; 2) According to the process, the deep learning algorithms are classified into direct detection algorithms and indirect detection algorithms, and the advantages and disadvantages of the current license plate detection algorithms and character recognition algorithms are analyzed; 3) The differences in data sets, workstation, accuracy and time of different license plate recognition systems are compared; 4) Compare and illustrate the existing public license plate datasets according to the number of pictures, resolution and environmental complexity, and make a prospect for the future research direction of license plate recognition. INDEX TERMS Image processing, image analysis, image classification, license plate recognition, deep learning.

show abstract

“…They have undertaken experiments on different angular orientations of license plate and reported accuracy and character recognition rate of their scheme. Xiang et al [20] have also proposed a method for license plate detection with the help of fully convolutional network and they found good accuracy rate, fact detection with low computational costs.…”

Section: B Earlier Work Donementioning

confidence: 99%

Feasibility of Efficient Number Plate Recognition using Morphological Dilation and Support Vector Machine

Vaishnav¹,

Ahuja²,

Mandot³

2019

IJRTE

View full text Add to dashboard Cite

Intelligent data acquisition of vehicle number plate plays a significant role to recognize a vehicle and it automatic parking, traffic movement and scheduling, tracking of stolen vehicle, and many more. Although different methodologies of automatic number plate reading have developed alongwith their algorithms, still an efficient number plate recognition technique for better segmentation and recognition of the captured number plate using Morphological Dilation and Support Vector Machine (SVM) are expected to be helpful. In this paper, we present a modified method for recognition of contents of number plate using morphological dilation and SVM. We have compared our results with those from the existing models using Wavelet Transform and Artificial Neural Network techniques. Superiority of present methodology is established using parameters like image segmentation and recognition.

show abstract

Lightweight fully convolutional network for license plate detection

Cited by 23 publications

References 17 publications

A Robust Real-Time Automatic License Plate Recognition Based on the YOLO Detector

A Robust Real-Time Automatic License Plate Recognition Based on the YOLO Detector

Research on License Plate Recognition Algorithms Based on Deep Learning in Complex Environment

Feasibility of Efficient Number Plate Recognition using Morphological Dilation and Support Vector Machine

Contact Info

Product

Resources

About