[ Pacific Security Network News ]
Vehicle License Recognition Technology (VPR) is an application of computer video image recognition technology in vehicle license plate recognition.
First, the license plate recognition implementation
The identification of the vehicle license plate is based on image segmentation and image recognition theory, and analyzes and processes the image containing the vehicle number plate to determine the position of the license plate in the image and further extract and recognize the text characters.
License plate recognition technology implementation and workflow
The identification steps are summarized as: license plate location, license plate extraction, and character recognition. The three-step identification work complements each other, and each has a higher efficiency, and the overall recognition rate will increase. The speed of recognition speed depends on character recognition. The main application technique of character recognition is to compare the sample library, that is, to build a sample library for all characters. After character extraction, the character is judged by comparing the sample library, and the recognition process will be generated. The intermediate result value such as credibility and inclination; the other is the character recognition technology based on the character structure knowledge, which is more effective in improving the recognition rate and accuracy, and has stronger adaptability.
There are two main ways to implement the license plate recognition system: one is the identification of still image images, and the other is the real-time recognition of dynamic video streams. The recognition efficiency of static image recognition technology is limited to the capture quality of images. For single-frame image recognition, the current market recognition speed is 200 milliseconds. The recognition technology of dynamic video stream is more adaptable and faster. Fast, it realizes the recognition of each frame of video, increases the number of recognition comparisons, and chooses the preferred license plate number. The key is that it is less affected by the quality of single-frame image. The current product identification time is 10 milliseconds.
License plate recognition is widely used in highway vehicle management. In the electronic toll collection (ETC) system, it is also the main means to identify the identity of vehicles in combination with DSRC technology.
In parking lot management, license plate recognition technology is also the main means of identifying vehicle identity. In the "Technical Requirements for Vehicle Image and Number Plate Information Collection and Transmission System of Parking Garage (Field)" constructed by Shenzhen Public Security Bureau, license plate recognition technology has become the main means of vehicle identification.
The license plate recognition technology is combined with the electronic non-stop toll collection system (ETC) to identify the vehicle. When the passing vehicle passes through the crossing, there is no need to stop, that is, the vehicle identity can be automatically recognized and automatically charged. In the management of the yard, in order to improve the traffic efficiency of the entry and exit vehicles, the license plate recognition is for vehicles that do not need to collect parking fees (such as monthly trucks, free internal vehicles), and build an unattended fast-track, free of access and non-stop access experience. Change the management mode of access to the parking lot.
Second, the license plate recognition technology workflow
The license plate recognition system uses a highly modular design that treats each part of the license plate recognition process as a separate module.
(1) Vehicle detection and tracking module
The vehicle detection and tracking module mainly analyzes the video stream, determines the position of the vehicle, tracks the vehicle in the image, and records the close-up picture of the vehicle at the best moment of the vehicle position. The system can be well received by adding the tracking module. The ground overcomes various external disturbances, so that a more reasonable recognition result can detect unlicensed vehicles and output the results.
(2) License plate positioning module
The license plate location module is a very important link and the basis of the follow-up link. Its accuracy has a great impact on the overall system performance. The license plate system completely abandons the previous algorithm and realizes a new license plate location algorithm based on learning multiple features, which is suitable for various complex background environments and different camera angles.
(3) License plate correction and fine positioning module
Due to the limitation of shooting conditions, the license plate in the image always has a certain inclination, and a correction and fine positioning link is needed to further improve the quality of the license plate image and prepare for the segmentation and recognition module. Using a well-designed fast image processing filter, not only is the calculation fast, but also the overall information of the license plate is used to avoid the influence of local noise. Another advantage of using this algorithm is that it can also fine-tune the license plate by analyzing multiple intermediate results, further reducing the impact of non-licence areas.
(4) License plate segmentation module
The license plate segmentation module of the license plate system utilizes various features such as grayscale, color, and edge distribution of the license plate text, which can better suppress the influence of other noises around the license plate and can tolerate license plates with a certain inclination angle. This algorithm is useful for applications such as mobile inspections where the license plate image is noisy.
(5) License plate recognition module
In the license plate recognition system, a combination of multiple recognition models is usually used to identify the license plate, and a hierarchical character recognition process is constructed, which can effectively improve the correct rate of character recognition. On the other hand, before the character recognition, the computer intelligent algorithm is used to pre-process the character image, which not only preserves the image information as much as possible, but also improves the image quality, improves the distinguishability of similar characters, and ensures the reliability of character recognition.
(6) License plate recognition result decision module
The recognition result decision module, specifically, the decision module utilizes the history records left by the process of the license plate passing through the field of view to make an intelligent decision on the recognition result. It obtains the comprehensive credibility evaluation of the license plate by calculating the number of observation frames, the stability of the recognition result, the trajectory stability, the speed stability, the average credibility and the similarity, thereby determining whether to continue tracking the license plate or output. Identify the result or reject the result. This method comprehensively utilizes the information of all frames, reduces the contingency error caused by the previous single image-based recognition algorithm, and greatly improves the recognition rate and the correctness and reliability of the recognition result.
(7) License plate tracking module
The license plate tracking module records various historical information such as the position, appearance, recognition result, credibility and the like of the license plate in each frame of the vehicle during driving. Since the license plate tracking module adopts a motion model with a certain fault tolerance and an updated model, those license plates that are short-term occlusion or instantaneously blurred can still be correctly tracked and predicted, and finally only one recognition result is output.
In the field of energy power, the improvement of efficiency and the extension of life span have very demanding requirements on material performance and structural design. Silicon carbide ceramic materials outperform most metallic materials in terms of specific stiffness, thermal shock resistance, friction and wear resistance, and thermal conductivity, but are limited by the difficulty of processing ceramic materials, which prevents their large-scale application.
FLK uses additive manufacturing technology to achieve a breakthrough in the manufacture of integrated complex components made of silicon carbide ceramics.
High strength, high rigidity, high stiffness, light density 280-3.05gcm3, low energy consumption, good thermal shock resistance, small coefficient of thermal expansion, 3D printed silicon carbide ceramic parts can extend its service life by more than 1 times than metal parts, reduce operating energy consumption, and meet the extreme working conditions applications.
Mining And Powder,Wheel Mining And Powder,Wheel Grader Mining And Powder,Graded Wheel Mining And Powder
Ningbo FLK Technology Co., Ltd. , https://www.flk-global.com