EDF‐LPR: a new encoder–decoder framework for license plate recognition

Gao, Fei; Cai, Yonghua; Ge, Yisu; Lu, Shufang

doi:10.1049/iet-its.2019.0253

Cited by 8 publications

(4 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Different convolutional layers have been used in this method. Gao et al [30] proposed an end-to-end network for plate detection and recognition based on encoder and decoder. The efficiency of traditional plate detection methods is affected by several factors including light intensity, shadow, and complex background.…”

Section: Wireless Communications and Mobile Computingmentioning

confidence: 99%

An End-to-End Deep Learning Approach for Plate Recognition in Intelligent Transportation Systems

Pirgazi

Kallehbasti

Sorkhi

2022

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

Accurate and fast recognition of license plates is one of the most important challenges in the field of license plate recognition systems. Due to the high frame rate of surveillance cameras, old license plate recognition systems cannot be used in real-time applications. On the other hand, the presence of natural and artificial noise and different light and weather conditions make the detection and recognition process of these systems challenging. In this paper, an end-to-end method for efficiently detecting and recognizing plates is presented. In the proposed method, vehicles are first detected using a single-shot detector- (SSD-) based deep learning model in the video frames and the input images. This will increase the speed and accuracy in identifying the location of the plate in the given images. Then, the location of the plate is identified using the proposed architecture based on convolutional networks. Finally, using a deep convolutional network and long short-term memory (LSTM), the characters related to the plate are recognized. An advantage of our method is that the proposed deep network is trained using different images with different qualities that leads to high performance in detecting and recognizing plates. Also, considering that in the proposed method the vehicles are first detected and then the plate is detected in the vehicle image, there is no limit in the number of identified plates. Moreover, plate detection in the vehicle rectangle, instead of the whole frame, speeds up our method. The proposed method is evaluated using several databases. The first part of the evaluation focuses on robustness and recognition speed. The proposed method has the accuracy of 100% for vehicle detection, 100% for plate detection, and 99.37% for character recognition. In the second part of evaluation, the proposed method is evaluated in terms of overall speed. The experimental results witness that the proposed method is capable of processing 30 frames per second without losing any data and also outperforms several methods proposed in recent years, in terms of time and accuracy.

show abstract

Section: Wireless Communications and Mobile Computingmentioning

confidence: 99%

An End-to-End Deep Learning Approach for Plate Recognition in Intelligent Transportation Systems

Pirgazi

Kallehbasti

Sorkhi

2022

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

show abstract

“…Generally, the adaptability of CNN‐based methods is relatively weak, and it is difficult to flexibly adapt to license plates of different lengths using a single model. To solve this problem, character‐level detection combined with sequence labelling is proposed in [9]. Similarly, the end‐to‐end method proposed in [5] detects each character using a regression net, and then recognises each character using a classification net.…”

Section: Related Workmentioning

confidence: 99%

“…However, these methods are environmentally sensitive and will bring propagation of error between segmentation and recognition. To overcome these problems, character‐level detection can be used to locate more precise character areas [2, 7–9]. Generally, these methods are time consuming and depend on character‐level annotations, the manual labelling of which is a heavy workload.…”

Section: Introductionmentioning

confidence: 99%

Efficient and unified license plate recognition via lightweight deep neural network

Qin¹,

Liu

2020

IET image process

View full text Add to dashboard Cite

In this study, the authors are interested in building a unified deep learning framework to solve the recognition problem of both single-line and double-line car license plates. Most existing methods are designed for single-line license plate recognition. For double-line cases, detection and segmentation are usually adopted firstly to locate each line of characters. These methods are usually environmentally sensitive and will bring propagation of error between segmentation and recognition. To solve the problems, the authors propose a unified method that can recognise both single-line and double-line license plates in an end-to-end way without line segmentation and character segmentation. Specifically, an improved lightweight convolutional neural network is used to extract features efficiently. Then, the multi-task learning strategy is used to simultaneously perform license plate classification and character recognition. Finally, recognition task is treated as sequence labelling problems, which are solved by connectionist temporal classification directly. Experimental results indicate that the proposed method significantly outperforms the previous state-of-the-art methods on both public datasets and the synthetic dataset.

show abstract

“…It is well known that the prime issues in the recognition stage are the effectiveness and efficiency of extracting the distinctive feature from the patterns. Most of the recognition algorithms for recognizing alphanumeric characters, binary and ideographs like Stroke analysis [5,6] requires a thinned character with one pixel width. However, obtaining character with one-pixel thickness and noise free is itself a challenge.…”

Section: Introductionmentioning

confidence: 99%

Efficient Thinning Algorithm for Malaysian Car Plate Character Recognition

Shitu

Abu-Bakar²,

Tahir³

et al. 2021

Elektrika

View full text Add to dashboard Cite

The thinning algorithm is one of the approaches of identifying each character printed on the car plate. Malaysian car plate characters appear in different character sizes, styles, customized printed characters etc. These variations contribute to difficulty in thinning successfully segmented and extracted license plate characters for recognition. To address these problems, an improved thinning operation for Malaysian car plate character recognition is proposed. In this algorithm, samples from segmented and extracted license plates are used for a thinning operation which is passed to Zhang-Suen thinning algorithm that could not guarantee one pixel thick and then to single pixelate algorithm that provides one pixel width of character for recognition. From the simulation, the result obtained has clearly proven to be the best for character recognition systems with least number of white pixels (777 pixels) and 0.26% redundant pixel left in the medial curve.

show abstract

EDF‐LPR: a new encoder–decoder framework for license plate recognition

Cited by 8 publications

References 52 publications

An End-to-End Deep Learning Approach for Plate Recognition in Intelligent Transportation Systems

An End-to-End Deep Learning Approach for Plate Recognition in Intelligent Transportation Systems

Efficient and unified license plate recognition via lightweight deep neural network

Efficient Thinning Algorithm for Malaysian Car Plate Character Recognition

Contact Info

Product

Resources

About