The recognition principle of license plate recognition system

Automatic license plate recognition is a pattern recognition technology that uses vehicle dynamic video or still images to automatically identify the license plate number and license plate color. The hardware infrastructure generally includes triggering equipment (monitoring whether the vehicle enters the field of vision), camera equipment, lighting equipment, image acquisition equipment, and processor (such as a computer) for identifying the license plate number. Its software core includes license plate location algorithm, license plate character segmentation algorithm and Optical character recognition algorithms. Certain license plate recognition systems also have the function of judging whether there is a car through a video image or not, which is called video vehicle detection. A complete license plate recognition system should include vehicle detection, image acquisition, license plate recognition and other parts (as shown in Figure 1). When the vehicle detection section detects the arrival of the vehicle, the image acquisition unit is triggered to collect the current video image. The license plate recognition unit processes the image, locates the license plate position, and then separates the characters in the license plate for identification, and then forms the license plate number to output.

Vehicle inspection

Vehicle detection can use buried coil detection, infrared detection, radar detection technology, video detection and other methods. The use of video detection can avoid damage to the road surface, no need for additional external inspection equipment, no need to correct the trigger position, cost savings, and is more suitable for mobile and portable applications.

The system performs video vehicle detection, requires high processing speed and adopts an excellent algorithm to achieve image acquisition and processing without dropping frames. If the processing speed is slow, it will result in dropped frames, making it impossible for the system to detect a vehicle with a fast traveling speed. It is also difficult to ensure that recognition processing is started at a position that is conducive to recognition and the system recognition rate is affected. Therefore, combining video vehicle detection with automatic license plate recognition has certain technical difficulties.

Number identification

For license plate recognition, the following basic steps are required:

1) License plate location, location of the license plate in the positioning picture;

2) The license plate character is divided and the characters in the license plate are separated;

3) Recognition of license plate characters, identifying the segmented characters, and finally composing the license plate number.

In the license plate recognition process, the identification color of the license plate is based on different algorithms, which may be realized in the above-mentioned different steps, and usually cooperate with the license plate recognition to verify each other.

License plate positioning

Under the natural environment, the background of the car image is complex and the illumination is uneven. How to accurately determine the license plate area in the natural background is the key to the entire recognition process. Firstly, a wide range of relevant search is performed on the collected video images to find a number of areas that meet the characteristics of the automobile license plate as candidate areas. Then these candidate areas are further analyzed and judged, and finally an optimal area is selected as the license plate area. Separate it from the image.

License plate character segmentation

After completing the positioning of the license plate area, the license plate area is divided into individual characters and then recognized. Character segmentation generally adopts vertical projection method. Since the projection of a character in the vertical direction necessarily takes place near the local minimum between characters or spaces within the character, this position should satisfy the character writing format, characters, size restrictions, and some other conditions of the license plate. Using vertical projection method has a good effect on character segmentation in car images under complex environments.

License plate character recognition

The methods mainly include template matching algorithm and artificial neural network algorithm. The template-based matching algorithm first binarizes the segmented characters and scales their size to the size of the template in the character database, then matches all the templates and selects the best match as the result. There are two kinds of algorithms based on the artificial neural network: one is to extract features from the characters first, and then use the obtained features to train the neural network distributor; the other method is to directly input the images into the network, and the features are automatically extracted by the network until Identify the result.

In practical applications, the recognition rate of the license plate recognition system is also closely related to the license plate quality and the shooting quality. The quality of the license plate can be affected by various factors such as rust, stains, peeling paint, faded fonts, blocked license plates, tilted license plates, bright reflective, multiple license plates, fake license plates, etc. The actual shooting process will also be subject to ambient brightness. , shooting methods, vehicle speed, and other factors. These influencing factors reduce the recognition rate of license plate recognition in different degrees. It is also the difficulty and challenge of the license plate recognition system. In order to improve the recognition rate, besides constantly improving the recognition algorithm, it should also find a way to overcome various lighting conditions, so that the captured image is most conducive to recognition.

License plate recognition PTZ camera captures license plate through light suppression screen, electronic shutter adjustment, wide dynamic function, etc.:

Strong light shielding: Based on the low-light color camera, through the function of the software, the brightest part of the image is blocked. In traffic monitoring, the headlights are generally blocked by strong light, so that the license plate can be captured more clearly. However, the biggest drawback of this camera is that the software is not clear, the brightest part of the image is unclear, it is possible to block the license plate number, and it is impossible to handle the capture of high-speed moving objects. At present, domestic cameras do more in the light shielding, the effect of all parties reflected.

Adjustable electronic shutter: For high-speed movement of objects to capture (the speed of the car on the highway is generally more than 70KM / hour), you can achieve clear capture by reducing the electronic shutter speed, usually through manual adjustment. However, the biggest problem with this method is that daytime and evening illumination are different. It is necessary to set two shutter speeds to adapt to daytime and nighttime monitoring respectively.

Wide dynamic function: This is the best way to solve the impact of the lights on the capture, wide dynamic was first proposed by the Panasonic company. When the background light is too bright, ordinary cameras cannot solve the display problem of light and dark images.

As the most important means of identifying vehicle identities in the field of intelligent transportation, license plate recognition technology has played a significant role in realizing the intelligent management of traffic, and it has infiltrated the license plate recognition technology in various works.

Monitoring alarm

For vehicles that are included in the "black list", such as vehicles that are wanted or reported missing, underpayment vehicles, vehicles that do not have annual inspections, accidents escaping and illegal vehicles, etc., simply enter their license plate number into the application system, and the license plate recognition equipment is installed. At designated junctions, bayonet or by law enforcement officers, they can be placed as required. The system will read all the vehicle's license plate numbers and compare them with the “black list” in the system. Once the designated vehicle is detected, it will immediately send an alarm message. The system can work uninterrupted throughout the day, without fatigue, with a very low error rate; it can adapt to high-speed vehicles; it can complete tasks in the vehicle without affecting normal traffic; the driver will not be aware of the entire monitoring process and high confidentiality. . Applying such a system will greatly increase the efficiency of law enforcement.

Speeding violation penalty

The license plate recognition technology combined with the speed measuring device can be used for speeding violations of vehicles and is generally used for expressways. The specific applications are: setting speed monitoring points on the road, capturing speeding vehicles and identifying license plate numbers, sending license plate numbers and pictures of illegal vehicles to various exits; setting penalties at each exit, using vehicle number plate recognition devices to identify passing vehicles and adding numbers Compare with the number of speeding vehicles that have been received. Once the numbers are the same, an alert device is activated to notify law enforcement officers. Compared with traditional speed monitoring methods, this application can save the police force, reduce the work intensity of law enforcement personnel, and it is safe, efficient, and concealed. The driver must always remind himself not to speed, which greatly reduces the accident caused by speeding.

Vehicle access management

The license plate recognition equipment is installed at the entrance and exit, record the license plate number and the time of entry and exit of the vehicle, and combine it with the control equipment of automatic doors and barriers to realize the automatic management of the vehicle. Applied to the parking lot can realize automatic time-based charging, and can also automatically calculate the number of available parking spaces and give hints. The automatic parking fee management can save manpower and improve efficiency. The application to the smart community can automatically determine whether the entry vehicle belongs to the local cell and implement automatic timing and charging for non-interior vehicles. In some units, this kind of application can also be combined with the vehicle dispatching system to automatically and objectively record the vehicle's outbound condition. H3 Smart uses the license plate recognition technology in the traffic management system to achieve no parking and exemption from cards. Improve the efficiency of vehicle access.

Automatic release

The designated license plate information is entered into the system, and the system automatically reads the license plate of the passing vehicle and queries the internal database. For vehicles that need to be automatically released to drive electronic gates or barriers to allow them to pass, warnings will be given to other vehicle systems and handled by on-duty personnel. Can be used for special units (such as military administrative areas, confidential units, key protection units, etc.), road and bridge charge bayonet, and advanced residential areas.

Highway toll management

License plate recognition equipment is installed at each entrance of the expressway. When the vehicle enters, the vehicle license plate is identified and the entrance data is stored in the toll collection system. When the vehicle reaches the exit, the license plate is identified again and the entrance information is called according to the license plate information. The entrance and exit data are used to realize the toll management. This application can realize automatic billing and can prevent cheating, thus avoiding the loss of receivables.

Expressway has begun to implement networking fees. With the expansion of the scope of networking, the difference between the price of different types of vehicles is also getting higher and higher. Drivers using existing loopholes in the system will use the loopholes in the existing toll collection system to avoid evading fees. License plate recognition technology is the fundamental method to solve such problems.

Calculate vehicle travel time

In the traffic management system, the average travel time of a vehicle on a road can be used as a parameter for judging the road congestion condition. The license plate recognition equipment is installed at the start and end points of the road, and all passing vehicles are read and the license plate number is transmitted back to the traffic command center. Based on these results, the command center management system can calculate the average travel time of the vehicle.

Automatic registration of license plate number

The traffic supervision department handles a large number of images of illegal vehicles every day. Usually, the number of the license plate is manually identified and then entered into the management system. This method has a large workload and is prone to fatigue and misjudgment. Using automatic identification can reduce the work intensity and can greatly improve the processing speed and efficiency. This feature can be used for electronic police systems, road monitoring systems, etc.

The license plate recognition car park management system automatically recognizes and converts the vehicle license plate number image captured by the camera at the entrance into a digital signal. To achieve one card and one vehicle, the advantage of license plate recognition is that it can associate the card with the car, so that the management can be improved by one grade. The corresponding advantages of the card and the car are that the long-term card must be used in conjunction with the car to eliminate the use of one card and multiple vehicles. Vulnerabilities improve the benefits of property management; at the same time, they automatically compare incoming and outgoing vehicles to prevent theft. The upgraded camera system can collect clearer pictures and save them as files, which can provide strong evidence for disputes. It facilitates management personnel to compare the vehicles when they appear on the market, which greatly enhances the security of the system.

License plate recognition has a wide application prospect

With the continuous development of parking management systems, license plate recognition technology has also begun to apply to parking lots. However, according to investigations, as only a few companies in China have specially developed license plate recognition technology for parking lots, many of the license plate recognition technologies used in current parking lots are copied from early license plate recognition technologies, and the two are in recognition environment and identification angles. The large differences in the lighting environment, vehicle speed, and other aspects have also led to the emergence of "acclimatization" in most of these license plate recognition systems. A lot of parking lot calls for "private custom" license plate recognition system.

According to the latest statistics, by the end of 2015, the market demand for China's parking lot will exceed 3.7 billion yuan, and the next five years will surely be the golden period for the intelligentization of parking lots. Automatic license plate recognition is the most valuable market in the future parking market. With the integration of high-definition and intelligence, license plate recognition can also help us solve more problems in our lives.

License plate recognition integrated machine has excellent imaging control, can automatically track light changes, effectively inhibit the smooth light and backlight, especially at night to suppress the interference of the car headlights, thus clearly capturing the license plate; and for the temporary absence of license plates or serious damage to the license plate The situation can also realize intelligent traffic management; in addition, offline operation is also a unique advantage of the license plate recognition integrated machine, which can effectively improve the efficiency of vehicle access.

In the future, with the acceleration of the development of China's urbanization process, traffic pressure will become more severe, so intelligent traffic management will be the general direction of future traffic development. As an important core in the intelligent traffic management system, the license plate recognition system will also be further supported and developed. Experts predict that the future license plate recognition technology will be more widely used, and the license plate recognition system industry will also face a major reshuffle. Only companies with independent core technologies and product quality standards can pass the inspection. This is also the process of license plate recognition technology moving toward rapid development. The only way to go.