Application of Automatic License Plate Recognition Technology in Intelligent Transportation System
chrispeng [ad_1]
The automatic license plate recognition system is an important part of intelligent transportation. Its main task is to detect and analyze the original image of the vehicle output by the acquisition unit, extract the relevant feature information of the license plate, compare and recognize it, and it can be used without any changes to the vehicle. In this case, effectively record and verify vehicle number plate information.
In recent years, automatic license plate recognition technology has been widely used in urban intelligent transportation systems, such as red light capture, speeding violation capture, and traffic security checkpoint system, especially the traffic security checkpoint system, which is an important technology for public security and criminal investigation management One of the methods is to put forward higher requirements on automatic license plate recognition technology, and promote the rapid development of automatic license plate recognition technology. The emergence of high-definition systems has greatly improved the image resolution and provided a good basic condition for automatic license plate recognition technology. The accuracy of license plate recognition has been greatly improved, and the corresponding recognition basic data has been continuously mined.
The bayonet system generally adopts a strategy based on distributed centralized management, and organically combines the front-end physical layer, transmission network layer, data processing layer and user application layer of the system through a multi-level three-dimensional structure.
The system is mainly composed of front-end data acquisition subsystem, network transmission subsystem, central management subsystem and other parts. The front-end data acquisition subsystem obtains the elapsed time, speed, picture, license plate number, body color and other data of the vehicle through video tracking and analysis technology. The data is transmitted to the central management subsystem through the network transmission subsystem. The central management subsystem performs centralized management, storage, and sharing of data.
Automatic license plate recognition process
The front end of the system adopts an embedded high-definition integrated camera, which can realize the output of video and picture stream with a resolution of one million. It has a built-in high-performance DSP chip, supports built-in intelligent algorithms, and can realize functions such as video detection and automatic license plate recognition. The automatic license plate recognition system with built-in camera uses a unique texture + model algorithm. It has the characteristics of accurate positioning, fast recognition, high recognition accuracy, and low misrecognition rate. It can not only capture vehicles with license plates, but also for unlicensed vehicles. Perform normal capture. The license plate recognition algorithm based on the back-end server or front-end industrial computer in the traditional model is transplanted to the front-end camera. It has the characteristics of high integration, high stability, and high adaptability. Compared with the traditional PC or industrial computer mode, it is more adaptable The complex environment of the actual road can better meet the requirements of all-weather work in the intelligent transportation system. The dynamic video recognition technology is adopted to realize the recognition of each frame of the video stream, so as to increase the number of recognition comparisons and greatly improve the efficiency and accuracy of recognition.
The automatic recognition of vehicle license plates is mainly based on image segmentation and image recognition theories to analyze and process images containing vehicle license plates to determine the position of the license plate in the image, and to further extract and recognize text characters. The specific steps of recognition are divided into license plate location, license plate extraction, and character recognition. In the natural environment, the camera first performs a large-scale related search on the collected video images, finds several areas that match the characteristics of the car license plate as candidate areas, and then further analyzes and judges these candidate areas, and finally selects the best one The area is used as the license plate area, and it is segmented from the image.
After the registration of the license plate area is completed, the license plate area is divided into individual characters and then recognized. The license plate recognition algorithm adopts a template-based matching algorithm. First, the divided characters are binarized and their size is scaled to the template in the character database Then, it is matched with all templates, and finally the best match is selected as the result. Through this multiple comparison method, the accuracy of license plate recognition is greatly improved.
License plate recognition function
The bayonet system adopts advanced image recognition algorithms to realize automatic recognition of vehicle number recognition, number plate color recognition, body color and car model of all vehicles passing through.
1. The number plate structure that can be recognized by the number plate structure recognition system includes
Number plates with single-line character structure, such as military small car number plates, small car number plates in GA36-2007, Hong Kong and Macau entry and exit number plates, coach car number plates, etc.;
Small car number plates for the armed police; police car number plates; number plates with platoon character structure, such as large-scale military car number plates, large-scale armed police car number plates, large car number plates in GA36-2007, trailer number plates, and low speed Car number plates, etc.
2. Number plate character recognition
The recognized characters include: numbers: 0~9; letters: A~Z; abbreviations of provinces, autonomous regions, and municipalities directly under the Central Government; Chinese characters used for classification of military license plates; special characters for armed police number plates: WJ, 00~34, practice.
3. Number plate color recognition
The system can identify motor vehicle license plates with four background colors: blue, yellow, white, and black. The system uses a combination of license plate color and video detection technology to classify vehicles. For civilian vehicles, blue license plates indicate small vehicles, while yellow license plates indicate large vehicles. Therefore, first use the color of the license plate to determine the type of vehicle. For situations where the vehicle type cannot be determined or the color of the license plate cannot be determined based on the color of the license plate, image analysis technology is used to assist in distinguishing the type of vehicle.
4. Vehicle number plate recognition
The number plate identification information includes information such as the number plate structure, the number plate characters, and the number plate color.
Application Features of Automatic License Plate Recognition System
1. Powerful ISP processing capabilities
The recognition rate of the automatic license plate recognition system is closely related to the quality of the license plate and the quality of the image taken. Not only the rust, stain, peeling paint, and font fading of the license plate itself will greatly affect the accuracy of license plate recognition, but also Whether the shooting environment is ideal will also have a great impact on license plate recognition.
The intelligent traffic camera has built-in powerful ISP processing functions, which can provide video stabilization, face detection, noise filtering, automatic white balance, automatic exposure and gamma correction, edge enhancement and other functions, which will improve the image quality and effect to a new level. It not only improves the user’s actual perception, but also provides a good calculation and analysis basis for more intelligent applications such as license plate recognition, which fully guarantees a higher accuracy of license plate recognition.
2. Good adaptability to light and climate background
Many license plate recognition systems have a higher recognition rate when it is cloudy, but decrease or even fail to recognize when it is sunny. In the case of direct light, the shooting direction is the same as the direction of sunlight. The license plate area is very bright, which leads to thicker strokes and adhesion of characters. Moreover, my country’s license plates are made of reflective paint. In severe cases, mirror reflections may appear and the license plate numbers cannot be seen clearly. . In addition, the bright lines and halos generated by the reflection on the surface of the car body will also affect the recognition. License plate recognition is mostly used to identify vehicles in motion. The license plate area is not fixed in the entire image, and ordinary cameras cannot adjust according to the license plate area. Vehicles turn on the lights in the night environment, and ordinary cameras are affected by the headlights and the exposure intensity is reduced. The license plate area of the image is very dark, and the number cannot be seen clearly. The light from the headlights may also form a large halo to block the license plate area.
The ambient light dynamic analysis technology and local brightness feedback closed-loop control technology can analyze and control the overall brightness of the image and the brightness of the license plate area in real time, and intelligently adjust the camera’s iris, shutter, gain and other exposure parameters to dynamically track changes in light. The complex environment, climate and light changes have good adaptability, and the clearest images can be taken all-weather, thus ensuring a very high license plate recognition rate.
3. Accurately capture vehicles without rear license plates or hidden by rear license plates
The system uses mature, stable and reliable ground sensing coils and advanced video analysis and detection technology. It can also accurately capture vehicles without a rear license plate or deliberately blocked the rear license plate in order to avoid capturing, ensuring the accuracy and completeness of the record of violations at intersections, and implementing it for the traffic control department Provide reliable clues and basis for traffic management measures, penalties for violations, etc.
4. Multi-lane, multi-vehicle number plate recognition at the same time
License plate recognition is not an isolated technology, but is closely integrated with practical applications. It fully considers the various actual conditions of the system application, and has been specially designed for the recognition of multiple lanes and multiple vehicles at the same time. Some manufacturers use the most advanced visual analysis algorithms, which not only can quickly locate all the license plates in the screen, quickly identify and output the recognition results in a very short time, but also ensure a very high license plate recognition rate.
5. Fast license plate recognition speed
The speed of license plate recognition determines whether the license plate recognition system can meet the requirements of real-time practical applications. A system with a high recognition rate, if it takes a few seconds or even a few minutes to recognize the result, the system will be meaningless because it cannot meet the real-time requirements in practical applications. The license plate recognition algorithm built in the intelligent traffic camera has extremely high recognition efficiency. The average time of single license plate recognition is about 40ms. The faster recognition speed can avoid the missing license plate recognition, and it can release more information for other intelligent analysis applications in time. More system resources.
6. License plate recognition pixels and high angle tolerance
The license plate recognition technology has high recognition pixels and angle tolerance, and the size of the recognition license plate can reach 75 pixels to 220 pixels; it supports the recognition of a certain degree of inclination of the license plate, and it can be recognized normally within ±15°. It has good adaptability to the size of the license plate, the distance of the vehicle appearing in the screen, and the skewed position, which greatly improves the practicability of the system.
The role of license plate recognition technology for traffic control
1. The use of license plate recognition technology will greatly reduce traffic violations and vicious traffic accidents. It will also provide strong evidence for various traffic accidents and the post-processing of life and property safety. It will play a pivotal role in my country’s traffic security and other aspects. No matter what trigger method is used, a mature license plate recognition system can effectively realize real-time monitoring and analysis of passing vehicles, and obtain various information such as license plate numbers, license plate colors, and vehicle types. Provide strong support for listing motor vehicles, investigating escaping vehicles in traffic accidents, analyzing traffic conditions, and strengthening public security management.
2. The intelligent transportation system based on license plate recognition can prevent the theft of motor vehicles, theft, counterfeit licenses, decking, smuggling, black market transactions and other increasingly rampant criminal activities in a timely manner. Through the installation and registration of the “electronic license plate” information of motor vehicles, the monitoring center can effectively remote control, grasp the image, digital information and traveling direction of the suspicious vehicle, and feedback the tracked information back to the monitoring center at any time. Based on this information, the public security department can understand, track, and control illegal vehicle transactions, vehicle theft, and other crimes in a timely manner. For counter-licensed and counter-licensed vehicles, the detection and recognition system will send out an alarm message when the electronic license plate number does not match the external license plate during the detection process, so that the public security department can pursue it.
3. The intelligent traffic management system based on license plate recognition can provide accurate and detailed classified traffic statistics data for urban road planning and design, realize the optimal design of road planning and management, and reduce traffic congestion black holes. The intelligent traffic management system can realize the sampling of vehicle traffic data at the city’s main road intersections, and perform data analysis on the types of vehicles (such as buses, trucks, buses, cars, taxis, etc.) and flow, and provide traffic flow for road planning and design Accurate data such as, vehicle type, peak period and peak value, scientifically guide road planning.
4. The use of an intelligent traffic management system based on license plate recognition can better solve various “old and difficult” problems faced in current traffic management.
Concluding remarks
After years of development, the automatic license plate recognition system has become a relatively mature technology. The traditional license plate recognition system is based on analog SD images for detection and recognition. Due to the low resolution of the SD images, the lack of layering, and the small field of view, the license plate recognition cannot achieve the desired effect, often in order to achieve the license plate recognition rate. It is necessary to sacrifice the panoramic view of the vehicle. Therefore, two cameras are required to complete the close-up of the license plate and the recording of the panoramic view of the vehicle. The system complexity is relatively high.
I believe that in the next few years, with the continuous application and construction of high-definition intelligent transportation systems in various places, automatic license plate recognition technology will gradually develop to high-definition, integrated, and intelligent development. It will continue to play its more and more important role in various application systems. The role of.
[ad_2]
