Vehicle identification method and device based on laser ranging

Abstract

The invention discloses a vehicle identification method and device based on laser ranging, and the method comprises the steps: controlling a laser sensor to detect at least one vehicle, and triggering a shooting control instruction when the vehicle enters a preset detection region of the laser sensor; the shooting control instruction controls a strobe light to generate a transient flash to irradiate the vehicle, and controls a camera to continuously shoot a plurality of image data of the vehicle; identifying the plurality of image data by using a vehicle identification neural network model, and outputting the vehicle type information, the license plate type and the license plate number of the vehicle; and storing the vehicle type information, the license plate type and the license plate number of the vehicle in a local database, and sending the vehicle type information, the license plate type and the license plate number of the vehicle to a cloud for storage. According to the invention, a reliable identification result can be realized under the condition of poor image quality; and storing and outputting the identified information such as the vehicle type and the license plate for further processing or recording by the system.

Description

Technical Field

[0001] This invention relates to the field of vehicle identification, and more particularly to a vehicle identification method and apparatus based on laser ranging. Background Technology

[0002] Currently, commonly used vehicle recognition technologies include image recognition and laser ranging. Common image recognition technologies, such as OCR, analyze vehicle images to obtain vehicle type and license plate information. However, recognition accuracy is limited at night, in rainy weather, or under poor lighting conditions. Disadvantages of existing technologies:

[0003] (1) Low recognition accuracy

[0004] Image recognition technology cannot accurately identify vehicle type and license plate information in complex environments, such as low-light conditions.

[0005] (2) The license plate is blurry.

[0006] When a vehicle is traveling at high speed or passing quickly, the license plate image is often blurry, making it impossible to correctly identify the license plate information.

[0007] (3) Inefficiency

[0008] Existing technologies are slow to identify large numbers of vehicles, failing to meet the demands for real-time monitoring and efficient processing. Summary of the Invention

[0009] To address the aforementioned technical problems, this invention provides a vehicle identification method and apparatus based on laser ranging.

[0010] A first aspect of the present invention provides a vehicle identification method based on laser ranging, comprising:

[0011] The laser sensor is controlled to detect at least one vehicle. When the vehicle enters the preset detection area of ​​the laser sensor, a shooting control command is triggered.

[0012] The shooting control command controls the strobe light to produce a brief flash to illuminate the vehicle, and controls the camera to capture multiple image data of the vehicle in succession.

[0013] The vehicle recognition neural network model is used to identify the multiple image data and output the vehicle model information, license plate type and license plate number;

[0014] The vehicle model information, license plate type, and license plate number are stored in a local database, and then sent to the cloud for storage.

[0015] In an optional embodiment, the shooting control command controls the strobe light and flash device to produce a brief flash to illuminate the vehicle, including:

[0016] When it is daytime, the shooting control command controls the strobe lights located in front of the vehicle to produce a brief flash;

[0017] When it is nighttime, the shooting control command controls the strobe lights located at the rear of the vehicle to produce a brief flash.

[0018] In an optional implementation, after recognizing the plurality of image data using a vehicle recognition neural network model, the method further includes:

[0019] Determine whether the license plate number in the multiple image data conforms to the license plate rules. If it does, output the vehicle model information, license plate type, and license plate number; otherwise, issue an alarm.

[0020] In an optional implementation, after determining whether the license plate number in the plurality of image data conforms to the license plate rules, the method further includes:

[0021] Determine whether the license plate numbers in the multiple image data are the same or whether the number of identical license plate numbers reaches a threshold. If so, output the vehicle model information, license plate type, and license plate number; otherwise, issue an alarm.

[0022] In one optional implementation, sending the vehicle model information, license plate type, and license plate number to cloud storage includes:

[0023] The vehicle model information, license plate type, and license plate number are sent to the cloud via MQTT or FTP protocols.

[0024] The method of using a vehicle recognition neural network model to identify the multiple image data includes:

[0025] The YOLOLPRNet vehicle recognition algorithm was run on the Jetson TX2 edge computing platform to identify the multiple image data.

[0026] A second aspect of the present invention provides a vehicle identification device based on laser ranging, comprising:

[0027] The laser detection module is used to trigger a shooting control command when the vehicle enters the preset detection area of ​​the laser sensor;

[0028] The control module is used to generate the shooting control command to control the strobe light and flash device to produce a short flash to illuminate the vehicle, and to control the camera to capture multiple image data of the vehicle in succession.

[0029] The recognition module is used to recognize the multiple image data using a vehicle recognition neural network model, and output the vehicle model information, license plate type and license plate number;

[0030] The storage module is used to store the vehicle model information, license plate type and license plate number in a local database, and to send the vehicle model information, license plate type and license plate number to cloud storage.

[0031] In one optional embodiment, the strobe light includes a front strobe light and a rear strobe light, and the camera includes a front camera and a rear camera. The front strobe light and the front camera are used to capture the front of the vehicle, and the rear strobe light and the rear camera are used to capture the rear of the vehicle. The front strobe light and the rear strobe light have opposite shooting directions, and the front camera and the rear camera have opposite shooting directions.

[0032] The control module is also used to: when it is daytime, control the strobe light located in front of the vehicle to produce a brief flash; when it is nighttime, control the strobe light located behind the vehicle to produce a brief flash.

[0033] A third aspect of the present invention provides a vehicle identification device based on laser ranging, comprising:

[0034] The controller, along with a laser rangefinder, camera, strobe light, and edge computing platform connected to the controller, allows the following actions to be taken: When the laser rangefinder detects a vehicle entering a preset detection area, the controller controls the strobe light to illuminate the vehicle and controls the camera to capture multiple images of the vehicle. The edge computing platform receives and identifies the multiple image data, and outputs the vehicle's model information, license plate type, and license plate number to the controller. The controller stores the vehicle's model information, license plate type, and license plate number in a local database and sends them to cloud storage.

[0035] A third aspect of the present invention provides an electronic device comprising:

[0036] At least one processor; and at least one memory communicatively connected to the processor, wherein the memory stores program instructions executable by the processor, and the processor invokes the program instructions to perform the method as described in the first aspect of the embodiments of the present invention.

[0037] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a computer, performs the method described in the first aspect of the embodiments of the present invention.

[0038] This invention combines multiple technologies such as laser ranging, strobe lights, and cameras to provide high-quality image data in complex environments; it utilizes a neural network module to extract features of vehicle models and license plates and performs accurate recognition, achieving reliable recognition results even with poor image quality; and it stores and outputs the recognized vehicle model, license plate, and other information for further processing or recording by the system. Attached Figure Description

[0039] Figure 1 This is a schematic diagram of a vehicle identification device based on laser ranging in an embodiment of the present invention.

[0040] Figure 2 This is a schematic flowchart of a vehicle identification method based on laser ranging in an embodiment of the present invention.

[0041] Figure 3 This is a flowchart illustrating another vehicle identification method based on laser ranging in an embodiment of the present invention.

[0042] Figure 4 This is a flowchart illustrating another vehicle identification method based on laser ranging in an embodiment of the present invention.

[0043] Figure 5 This is a schematic diagram of a vehicle identification device based on laser ranging in an embodiment of the present invention. Detailed Implementation

[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0045] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0046] Please see Figure 1 This invention discloses a vehicle recognition device based on laser ranging, comprising: a controller, and a laser ranging sensor, a camera, a strobe light, and an edge computing platform connected to the controller. The strobe light provides sufficient illumination, offering a better shooting environment for the camera in low light conditions, rainy weather, or other situations with insufficient lighting. The laser ranging sensor, strobe light, and camera are mounted at appropriate locations on a support column. The combined application of the laser ranging sensor, strobe light, and camera provides high-quality image data acquisition in complex environments, increasing the accuracy of vehicle model and license plate recognition and solving the recognition problems of existing technologies in low-light or adverse environments.

[0047] When the laser rangefinder sensor detects a vehicle entering its preset detection area, the controller activates the strobe light to illuminate the vehicle and the camera to capture multiple images of the vehicle. For example, first, coordinates are used to determine the vehicle's detection area, which is also the camera's shooting area. This ensures the laser sensor's field of view covers the entire detection area and can accurately measure the vehicle's distance. Then, the laser sensor's measurement parameters, such as measurement range, accuracy, and measurement interval, are set according to specific needs. Next, the laser sensor is activated and periodically performs distance measurements. The laser sensor sends a laser beam and measures its return time or phase difference to calculate the distance between the vehicle and the sensor. A threshold is set; when the distance between the vehicle and the sensor is less than this threshold, the vehicle is considered to have entered the detection area. Then, using burst shooting technology, five images of the vehicle are quickly captured after the strobe light flashes.

[0048] The camera captures multiple images in succession and transmits them to the edge computing platform. The edge computing platform receives and identifies the multiple image data, and outputs the vehicle model information, license plate type, and license plate number to the controller. This invention uses the Jetson TX2 edge computing platform, which is very small, similar in size to a typical set-top box, and operates stably, allowing for direct integration into the camera. Running the YOLOLPRNet vehicle recognition algorithm within this edge computing platform achieves an accuracy rate of over 98% for license plate and vehicle model recognition, significantly higher than traditional image recognition algorithms.

[0049] The controller stores the vehicle model information, license plate type, and license plate number in a local database, and sends the vehicle model information, license plate type, and license plate number to the cloud storage, so that it can be provided to relevant departments or systems for further processing or recording, such as speeding violations. It can be easily integrated into existing traffic management and safety monitoring systems to provide comprehensive vehicle identification and information collection functions.

[0050] This invention employs laser ranging and rapid image acquisition technology to achieve rapid vehicle detection and continuous shooting, enabling the acquisition of a large amount of vehicle image data in a short time, thus improving processing efficiency; it also solves the problem of blurred license plates when driving at high speeds or passing quickly, ensuring the reliability of the recognition results.

[0051] Please see Figure 2 As shown, the present invention also provides a vehicle identification method based on laser ranging, comprising the following steps:

[0052] Step 201: Control the laser sensor to detect at least one vehicle. When a vehicle enters the preset detection area of ​​the laser sensor, a shooting control command is triggered. The laser sensor emits a laser beam. When the laser beam encounters a vehicle, the reflected laser light is received by a receiver, and the distance between the vehicle and the sensor is measured. Typically, the distance to the laser sensor is the camera's shooting area, where the camera captures the clearest image.

[0053] Step 202: The shooting control command controls the strobe light to produce a brief flash to illuminate the vehicle, and controls the camera to capture multiple images of the vehicle in rapid succession. When the laser sensor detects the vehicle entering the detection area, the strobe light is triggered. The strobe light produces a brief flash to improve the quality of subsequent image acquisition. After the strobe light flashes, five images of the vehicle are quickly acquired using continuous shooting technology. The camera may include one or more cameras to capture different angles and views of the vehicle.

[0054] Step 203: Utilize a vehicle recognition neural network model to identify the multiple image data and output the vehicle model information, license plate type, and license plate number. The acquired images are analyzed and processed using the YOLOLPRNet vehicle recognition algorithm, employing techniques such as feature extraction, image recognition, and character recognition. This enables the simultaneous identification of multiple vehicles. First, deep learning-based object detection technology is used to detect the vehicle model, license plate type, and license plate region. Then, through image cropping, the license plate region is input into a deep learning-based end-to-end detection network for license plate number recognition. Finally, the correct vehicle model, license plate type, and license plate number information are output.

[0055] Step 204: Store the vehicle model information, license plate type, and license plate number in a local database, and then send these information to the cloud for storage. The identified vehicle model, license plate type, and license plate number can be stored in the local database and sent to the cloud via the MQTT protocol for subsequent querying and analysis.

[0056] Further, in step 202, the shooting control command controls the strobe light and flash device to generate a brief flash to illuminate the vehicle, including:

[0057] During the daytime, the shooting control command controls the strobe lights located in front of the vehicle to produce a brief flash; at night, the shooting control command controls the strobe lights located behind the vehicle to produce a brief flash.

[0058] Combination Figure 1 As shown, to prevent the strobe light from affecting the driver, a camera and strobe light located in front of the vehicle can be used when filming the vehicle during the day, while a camera and strobe light located behind the vehicle can be used when filming the vehicle at night.

[0059] Specifically, the strobe lights include a front strobe light and a rear strobe light, and the cameras include a front camera and a rear camera. The front strobe light and the front camera are used to capture images of the front of the vehicle, and the rear strobe light and the rear camera are used to capture images of the rear of the vehicle. The front and rear strobe lights have opposite shooting directions, and the front and rear cameras also have opposite shooting directions. It should be understood that the strobe lights and cameras are respectively installed on the support column in the forward and rearward directions.

[0060] Furthermore, such as Figure 3 As shown, in step 203 above, after recognizing the multiple image data using the vehicle recognition neural network model, the method further includes:

[0061] The system determines whether the license plate numbers in the multiple image data conform to license plate rules. If they do, it outputs the vehicle model information, license plate type, and license plate number; otherwise, it issues an alarm.

[0062] Determine whether the license plate numbers in the multiple image data are the same or whether the number of identical license plate numbers reaches a threshold. If so, output the vehicle model information, license plate type, and license plate number; otherwise, issue an alarm.

[0063] License plate numbers consist of six letters or numbers: The first letter represents the abbreviation of the province, autonomous region, or special municipality, such as Shanghai, Hebei, Yunnan, and Henan. The second letter or number represents the abbreviation of the city or district, such as H for Shanghai, J for Jinan, 01 for Dongcheng District, 11 for Huangpu District, 45 for Xi'an, and 02 for Hedong District. The last four letters or numbers are the vehicle's unique identifier, consisting of two letters or numbers followed by two letters, numbers, and Chinese characters, such as AA123X or VV 345Y. However, Tibet uses Arabic numerals, such as 01D 123. For private cars, the second letter or number in the license plate changes after signing agreements with Beijing, Tianjin, and Shanghai, such as H or H1 used in Xujiahui and Pudong New Area. Military vehicle license plates consist of three numbers and two letters, such as 870X, assigned by the Army vehicle management department. This also includes license plates for police vehicles, consulates, and other foreign institutions.

[0064] The starting point laser rangefinder detects the vehicle's arrival in the shooting area, and the camera takes five consecutive photos. The YOLOLPRNet vehicle recognition algorithm is used to identify the vehicle's model information, license plate type, and license plate number. If the license plate number and type conform to the license plate rules, it is checked whether all five recognized license plate numbers are the same. If they are all the same, the vehicle is considered correct. If only one of the license plate numbers is different from the others, it is also considered correct. If two or more of the license plate numbers are different from the other recognized results, the recognition is considered incorrect, and the recognition process continues. If the vehicle cannot be recognized, an error is reported.

[0065] The recognition results can be stored locally or uploaded to the cloud. Specifically, the vehicle model information, license plate type, and license plate number are sent to cloud storage. (See reference [link / reference]). Figure 4 As shown, the vehicle model information, license plate type, and license plate number are sent to the cloud via the MQTT or FTP protocol.

[0066] When relevant departments or systems conduct further processing or query records, the cloud server sends FTP file information to the requesting end. Through the WebSockets and MQTT protocols, the requesting end implements dynamic display using HTML5+JS front-end. Users can view the passing time, vehicle model, license plate number, and license plate type of the photographed vehicle in the vehicle information display.

[0067] Please see Figure 5 The present invention also provides a vehicle identification device based on laser ranging, comprising:

[0068] The laser detection module 51 is used to trigger a shooting control command when the vehicle enters the preset detection area of ​​the laser sensor;

[0069] The control module 52 is used to generate the shooting control command to control the strobe light and flash device to produce a short flash to illuminate the vehicle, and to control the camera to capture multiple image data of the vehicle in succession.

[0070] The recognition module 53 is used to recognize the multiple image data using a vehicle recognition neural network model, and output the vehicle model information, license plate type and license plate number;

[0071] The storage module 54 is used to store the vehicle model information, license plate type and license plate number in a local database, and to send the vehicle model information, license plate type and license plate number to cloud storage.

[0072] The strobe light includes a front strobe light and a rear strobe light, and the camera includes a front camera and a rear camera. The front strobe light and the front camera are used to photograph the front of the vehicle, and the rear strobe light and the rear camera are used to photograph the rear of the vehicle. The front strobe light and the rear strobe light shoot in opposite directions, and the front camera and the rear camera shoot in opposite directions.

[0073] The control module 52 is further configured to: when in daytime, control the shooting control command to control the strobe light located in front of the vehicle to produce a brief flash; when in nighttime, control the shooting control command to control the strobe light located behind the vehicle to produce a brief flash.

[0074] The above description of the vehicle recognition device based on laser ranging can be found in the description of the vehicle recognition method based on laser ranging, and will not be repeated here.

[0075] The present invention also provides an electronic device, comprising:

[0076] At least one processor; and at least one memory communicatively connected to the processor, wherein the memory stores program instructions executable by the processor, and the processor can execute the above-described laser ranging-based vehicle identification method by invoking the program instructions.

[0077] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described vehicle identification method based on laser ranging.

[0078] It is understood that computer-readable storage media can include: any entity or device capable of carrying computer programs, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), and software distribution media, etc. Computer programs include computer program code. Computer program code can be in the form of source code, object code, executable files, or certain intermediate forms, etc. Computer-readable storage media can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), and software distribution media, etc.

[0079] In some embodiments of the present invention, the device may include a controller, which is a microcontroller chip integrating a processor, memory, communication module, etc. The processor may refer to the processor included in the controller. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0080] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which embodiments of the invention pertain.

[0081] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0082] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A vehicle recognition method based on laser ranging, characterized by, The method comprises the following steps: controlling the laser sensor to detect at least one vehicle, and triggering a shooting control instruction when the vehicle enters a preset detection area of the laser sensor; the shooting control instruction controls the strobe light to generate a short flash to irradiate the vehicle, and controls the camera to continuously shoot multiple image data of the vehicle; using a vehicle recognition neural network model to recognize the multiple image data, and outputting vehicle model information, license plate type and license plate number of the vehicle; storing the vehicle model information, license plate type and license plate number in a local database, and sending the vehicle model information, license plate type and license plate number to a cloud storage. 2.The vehicle identification method based on laser ranging according to claim 1, characterized in that, The shooting control instruction controls the strobe light and the flash device to generate a short flash to irradiate the vehicle, which comprises: when in daytime, the shooting control instruction controls the strobe light located in front of the vehicle to generate a short flash; when in nighttime, the shooting control instruction controls the strobe light located behind the vehicle to generate a short flash. 3.The laser ranging based vehicle identification method of claim 1, wherein, After using the vehicle recognition neural network model to recognize the multiple image data, the method further comprises the following steps: judging whether the license plate number in the multiple image data conforms to the license plate rule, and if yes, outputting the vehicle model information, license plate type and license plate number; if not, performing an alarm. 4.The vehicle identification method based on laser ranging according to claim 3, characterized in that, After judging whether the license plate number in the multiple image data conforms to the license plate rule, the method further comprises the following steps: judging whether the license plate numbers in the multiple image data are the same or the number of the same license plate numbers reaches a threshold, and if yes, outputting the vehicle model information, license plate type and license plate number; if not, performing an alarm. 5.The laser ranging based vehicle identification method of claim 1, wherein, The method of sending the vehicle model information, license plate type and license plate number to the cloud storage comprises the following steps: sending the vehicle model information, license plate type and license plate number to the cloud through MQTT protocol or FTP protocol; The method of using the vehicle recognition neural network model to recognize the multiple image data comprises the following steps: running YOLOLPRNet vehicle recognition algorithm on Jetson tx2 edge computing platform to recognize the multiple image data.

6. A vehicle recognition device based on laser ranging, characterized by, The method comprises the following steps: a laser detection module for triggering a shooting control instruction when a vehicle enters a preset detection area of the laser sensor; a control module for generating the shooting control instruction to control the strobe light and the flash device to generate a short flash to irradiate the vehicle, and to control the camera to continuously shoot multiple image data of the vehicle; an identification module for using a vehicle recognition neural network model to recognize the multiple image data, and outputting vehicle model information, license plate type and license plate number of the vehicle; a storage module for storing the vehicle model information, license plate type and license plate number in a local database, and sending the vehicle model information, license plate type and license plate number to a cloud storage.

7. The laser ranging based vehicle identification apparatus according to claim 6, wherein The strobe light comprises a front strobe light and a rear strobe light, and the camera comprises a front camera and a rear camera; the front strobe light and the front camera are used to shoot the front of the vehicle, the rear strobe light and the rear camera are used to shoot the rear of the vehicle, the shooting direction of the front strobe light is opposite to that of the rear strobe light, and the shooting direction of the front camera is opposite to that of the rear camera. The control module is further configured to: when in the daytime state, the shooting control instruction controls the strobe light located in front of the vehicle to produce a short flash; and when in the nighttime state, the shooting control instruction controls the strobe light located behind the vehicle to produce a short flash.

8. A vehicle recognition device based on laser ranging, characterized by, The method comprises the steps of: The controller, the laser ranging sensor, the camera, the strobe light, and the edge computing platform are connected to the controller. When the laser ranging sensor detects that a vehicle enters a preset detection area of the laser sensor, the controller controls the strobe light to irradiate the vehicle and controls the camera to continuously shoot multiple image data of the vehicle. The edge computing platform receives the multiple image data and identifies the vehicle model information, the license plate type, and the license plate number of the vehicle, and outputs the vehicle model information, the license plate type, and the license plate number of the vehicle to the controller. The controller stores the vehicle model information, the license plate type, and the license plate number of the vehicle in a local database, and sends the vehicle model information, the license plate type, and the license plate number of the vehicle to a cloud storage.

9. An electronic device, comprising: The method comprises the steps of: at least one processor; and at least one memory in communication with the processor, wherein: the memory stores program instructions executable by the processor, and the processor invoking the program instructions can execute the laser ranging-based vehicle identification method according to any one of claims 1 to 5.

10. A computer-readable storage medium, characterized in that, A computer program is stored on the computer, and when the computer program is run by the computer, the laser ranging-based vehicle identification method according to any one of claims 1 to 5 is executed.

Images