The process of detecting vehicles’ license plates, along with recognizing the characters inside them, has always been a challenging issue due to various conditions. These conditions include different weather and illumination, inevitable data acquisition noises, and some other challenging scenarios like the demand for real-time performance in state-of-the-art Intelligent Transportation Systems (ITS) applications. This paper proposes a method for vehicle License Plates Detection (LPD) and Character Recognition (CR) as a unified application that presents significant accuracy and real-time performance. The mentioned system is designed for Iranian vehicle license plates, which have the characteristics of different resolution and layouts, scarce digits/characters, various background colors, and different font sizes. In this regard, the system uses a separate fine-tuned You Only Look Once (YOLO) version 3 platform for each of the mentioned phases and extracts Persian characters from input images in two automatic steps. For training and testing stages, a wide range of vehicle images in different challenging and straightforward conditions have been collected from practical systems installed as surveillance applications. Experimental results show an end-to-end accuracy of 95.05% on 5719 images. The test data included both color and grayscale images containing the vehicles with different distances and shooting angles with various brightness and resolution. Additionally, the system can perform the LPD and CR tasks in an average of 119.73 milliseconds for real life data, which illustrates a real-time performance for the system and usable applicability. The system is fully automated, and no pre-processing, calibration or configuration procedures are needed.

A Robust Deep Learning Approach for Automatic Iranian Vehicle License Plate Detection and Recognition for Surveillance Systems

Most existing methods for automatic license plate recognition (ALPR) focus on a specific license plate (LP) type, but little work focuses on multiple or mixed LPs. This article proposes a single neural network called ALPRNet for detection and recognition of mixed style LPs. In ALPRNet, two fully convolutional one-stage object detectors are used to detect and classify LPs and characters simultaneously, which are followed by an assembly module to output the LP strings. ALPRNet treats LP and character equally, object detectors directly output bounding boxes of LPs and characters with corresponding labels, so they avoid the recurrent neural network (RNN) branches of optical character recognition (OCR) of the existing recognition approaches. We evaluate ALPRNet on a mixed LP style dataset and two datasets with single LP style, the experimental results show that the proposed network achieves state-of-the-art results with a simple one-stage network.

/pdf/a-single-neural-network-for-mixed-style-license-plate-25tlargtye.pdf

A Single Neural Network for Mixed Style License Plate Detection and Recognition

In this paper, we propose a real-time and accurate automatic license plate recognition (ALPR) approach. Our study illustrates the outstanding design of ALPR with four insights: (1) the resampling-based cascaded framework is beneficial to both speed and accuracy; (2) the highly efficient license plate recognition should abundant additional character segmentation and recurrent neural network (RNN), but adopt a plain convolutional neural network (CNN); (3) in the case of CNN, taking advantage of vertex information on license plates improves the recognition performance; and (4) the weight-sharing character classifier addresses the lack of training images in small-scale datasets. Based on these insights, we propose a novel ALPR approach, termed VSNet. Specifically, VSNet includes two CNNs, i.e., VertexNet for license plate detection and SCR-Net for license plate recognition, integrated in a resampling-based cascaded manner. In VertexNet, we propose an efficient integration block to extract the spatial features of license plates. With vertex supervisory information, we propose a vertex-estimation branch in VertexNet such that license plates can be rectified as the input images of SCR-Net. In SCR-Net, we introduce a horizontal encoding technique for left-to-right feature extraction and propose a weight-sharing classifier for character recognition. Experimental results show that the proposed VSNet outperforms state-of-the-art methods by more than 50% relative improvement on error rate, achieving >99% recognition accuracy on CCPD and AOLP datasets with 149 FPS inference speed. Moreover, our method illustrates an outstanding generalization capability when evaluated on the unseen PKUData and CLPD datasets.

Rethinking and Designing a High-Performing Automatic License Plate Recognition Approach

In today’s modern era, parking remains a big problem for a lot of people. This problem consumes a person individual’s time by finding the right spot for parking. In the current research, the concept of an automatic parking system using the vehicle license plate or the number plate recognition is discussed. It will improve the process with much less hassle by removing human interaction. It will also lead to advancement in the security of vehicles evading the requirement of a slip or a magnetic card which is used to goes in and out for registering vehicles in a parking place. The researcher uses image processing algorithms to make an entry in the database of the parking automatically. AVNPR (Automatic Vehicle Number Plate Recognition) is used for the identification of the number of plates. Due to noise issues, deep algorithms like CNN (convolutional neural networks), RNN (recurrent neural networks) do not correctly recognize the miss-identification of the numbers in the vehicle plate. This problem is rectified by the authors by using the GAN (Generative adversarial networks) algorithm. GAN helps to create high-resolution images from a single low-resolution image. After applying the GAN, the classification of the vehicle plate is done through CNN. During experimentation, the proposed approach achieves 99.39% recognition accuracy for a vehicle number plate. Hence, the proposed system is suitable for identifying the numbers in the vehicle number plate automatically. Moreover proposed system compared with existing models, it has been found that it has achieved higher accuracy than the other models.

GAN-based synthetic data augmentation for increased CNN performance in Vehicle Number Plate Recognition

Automatic license plate recognition has a wide range of applications in intelligent transportation systems and is of great significance. However, most of the current work on license plate recognition focuses on the images on the front of license plates. license plate recognition in natural scenes and arbitrary perspective is still a huge challenge. To solve this problem, this work mainly studies the detection and recognition of inclined Chinese license plates in natural scenes. We propose a robust method that can detect and correct multiple license plates with severe distortion or skewing in one image and input them into the license plate recognition module to obtain the final result. Different from the existing methods of license plate detection and recognition, our method performs affine transformation during license plate detection to rectify the distorted license plate image. It can not only avoid the accumulation of intermediate errors but also improve the accuracy of recognition. As an additional contribution, we put forward a challenging Chinese license plate recognition data set, including images obtained from different scenes under a variety of weather conditions. Through a large number of comparative experiments, we have proved the effectiveness of our proposed method.

/pdf/robust-automatic-recognition-of-chinese-license-plates-in-1wpf1408tt.pdf

Robust Automatic Recognition of Chinese License Plates in Natural Scenes

License Plate Recognition (LPR) is of great significance due to its wide range of applications in the Intelligent Transportation System (ITS). It is an important and challenging research topic in image recognition fields. However, many of the current methods are still not robust in real-world complex scenario. The main contribution of this paper is to propose a multi-task convolutional neural network for license plate detection and recognition (MTLPR) with better accuracy and lower computational cost, and introduce a comprehensive data set of Chinese license plate. First, we train a Multi-task Convolutional Neural Networks (MTCNN) to detect license plate. Then we introduce an end-to-end method to recognize license plate information, which further improves the recognition precision. Last, We compare the experimental result with other state-of-the-art methods. The experimental result shows that our method achieves up to 98% recognition precision and is superior to other methods in the precision and speed of detection and recognition.

/pdf/a-light-cnn-for-end-to-end-car-license-plates-detection-and-4ylw9go3th.pdf

A Light CNN for End-to-End Car License Plates Detection and Recognition

The white paper of Aircraft tracking operation-Global Aeronautical Distress and Safety System (GADSS) is proposed by the International Civil Aviation Organization (ICAO) after MH370 event, which requires all member states to improve global tracking and monitoring capability of aircraft. It is difficult to realize the global tracking and monitoring of aircraft using conventional single surveillance method under the existing mature technology conditions. For example, there is the existence of coverage blind area using PSR and ADS-B due to the limitation of constructing ground station, and it is unknown the status of the aircraft operating in foreign countries because of the above data having not been shared all over the world. It is lack of timeliness and effectiveness for tracking and monitoring the aircraft owing to the low update rate, message loss, message false of the OOOI and positon message of ACARS. Different information sources(SSR, ADS-B, OOOI and positon message of ACARS, flight plan) for the same aircraft are associated by flight number, take-off/landing time and airport, also extrapolation criterion is used in this paper. SSR and ADS-B have priority to be used to surveillance the aircraft if above signals are covered. The remaining oil, take-off/landing time and airport, aircraft nationality registration numbers and the other important information are obtained by OOOI and positon message of ACARS and flight plan. ACARS is used to surveillance the aircraft without PSR or ADS-B. Otherwise, the flight plan is used to predict and deduce when the above three kinds of information are absent. The global tracking and monitoring of the whole flight process of the same flight is realized through complementary advantages of various information. Finally, all aircraft operating at home and abroad but attached to China can be tracked and monitored well, which ensured flight safety.

Research on global tracking and monitoring technology of aircraft based on multi-information sources

The playback of flights' surveillance information plays an import role when flights encounter special circumstances, such as accidents. It can help analyze when, where and why the special circumstances happened intuitively. Different from traditional air traffic control automation system, we propose a playback system based on double-buffers method. A playing buffer and a fetching buffer is created, one is for storing the surveillance information that will be play back and the other one is used for storing the surveillance information fetched from the database. Thus, data fetching and data playback can be done concurrently, to play back the flights' surveillance information continuously. What's more a reliable UDP is proposed to improve the reliability of the data communication of the playback system.

A playback system for flight surveillance information

Civil Aviation Administration of China (CAAC) proposed to enhance the global tracking and monitoring capability of China civil aircraft after the MH370 incident. In order to meet the urgent needs of global tracking for civil aircraft in China, the design scheme of flight data processing software for global flight tracking system was presented in this paper. Based on the features and operational process of global flight tracking, the overall framework and interface design of the software were given in detail. In order to meet the performance needs of the real-time processing of emergency mass data, the database access technology based on stored procedure was adopted and the realization method of flight data processing software was given. The engineering application and the test results shows that the software can well meet the functionality and performance requirements for the flight data processing of global flight tracking system, which provides technical reference for improving the global tracking and monitoring capability of China civil aircraft.

Jun Yang

Papers

A Light CNN for End-to-End Car License Plates Detection and Recognition

Research on global tracking and monitoring technology of aircraft based on multi-information sources

A playback system for flight surveillance information

Design and Implementation of Flight Data Processing Software for Global Flight Tracking System Based on Stored Procedure