Vehicle shooting method and related equipment

Abstract

The invention discloses a vehicle shooting method and related equipment, and the method comprises the steps: obtaining image data which comprises images of driving vehicles, determining whether there is an accent light vehicle whose vehicle lamp brightness is greater than a first threshold value in the driving vehicles or not based on the image data, if yes, determining shooting parameters and light supplement parameters of the accent light vehicle, and if not, determining the shooting parameters and the light supplement parameters of the accent light vehicle; and an image of the accent light vehicle is generated based on the shooting parameter and the light supplementing parameter, so that clear imaging of the accent light vehicle is realized.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of photographing technology, and specifically to a vehicle photographing method and related equipment. Background Art

[0002] With the rapid development of camera technology, cameras are being used more and more widely. For example, in traffic capture camera systems, it is necessary to realize video-based situation capture and image-based license plate capture.

[0003] On highways, expressways and other scenes, it is common to see some vehicles equipped with very bright strong lights near the license plate. The strong lights are not factory configurations of the vehicles, have high luminous intensity and dazzling light spots.

[0004] Because the strong light is installed close to the license plate, the light spot easily spreads to the license plate area, resulting in the general traffic capture camera system being unable to clearly image the license plate and unable to manage traffic behavior based on the license plate information. Summary of the invention

[0005] The embodiment of the present application provides a vehicle shooting method for achieving clear imaging of a vehicle with a strong headlight. The embodiment of the present application also provides a corresponding camera system, a computing device, and a computer-readable storage medium.

[0006] A first aspect of the present application provides a vehicle shooting method, the method comprising: acquiring image data, the image data comprising an image of at least one moving vehicle; determining whether there is a vehicle with high-spot lights among the at least one moving vehicle based on the image data, the headlight brightness of the vehicle with high-spot lights being greater than a first threshold; if there is a vehicle with high-spot lights among the at least one moving vehicle, determining a first shooting parameter and a first fill light parameter of the vehicle with high-spot lights; and generating a first image of the vehicle with high-spot lights based on the first shooting parameter and the first fill light parameter.

[0007] In the present application, the vehicle shooting method is applied to a camera system, and the camera system shoots at a road checkpoint to obtain image data, and the image data includes an image of at least one moving vehicle, that is, in the scene of the road checkpoint, if there is at least one moving vehicle, the camera system shoots and obtains image data. Alternatively, the camera system continuously shoots, and only when there is at least one moving vehicle does it obtain or save image data for subsequent processing.

[0008] The brightness of the headlights of the vehicles with strong headlights in this application is greater than the first threshold value, that is, the vehicles with strong headlights are vehicles that are equipped with and have turned on strong headlights. The strong headlights are not the factory configuration of the vehicle, the luminous intensity is high, the brightness is greater than the first threshold value, and the light spot is dazzling. Because the luminous intensity of the strong headlights is relatively consistent, generally higher than that of normal headlights, the license plate area can be exposed under relatively certain fill light conditions and exposure amounts to prevent overexposure or underexposure of the license plate. When the strong headlights are imaged, the light spots have similar shapes, and the brightness gradually decreases from the center to the surrounding areas, and most of them are elliptical. The relative position of the strong headlights on the moving vehicles is relatively fixed, mostly located in the central area below the front of the vehicle, adjacent to the license plate area, so the strong headlights can be detected through deep learning and other methods, and the hardware cost is reduced.

[0009] In the present application, after determining the existence and position of a vehicle with strong headlights, the first shooting parameters and the first fill light parameters for capturing the vehicle with strong headlights can be called up simultaneously, and the first shooting parameters and the first fill light parameters can be used to capture and generate a first image, which can clearly show the vehicle with strong headlights.

[0010] The first aspect is to acquire image data including images of moving vehicles, and then determine whether there are any vehicles with strong headlights whose headlight brightness is greater than a first threshold value among the moving vehicles based on the image data; if so, determine the shooting parameters and fill light parameters of the vehicle with strong headlights, and generate an image of the vehicle with strong headlights based on the shooting parameters and fill light parameters, thereby achieving clear imaging of the vehicle with strong headlights.

[0011] In a possible implementation of the first aspect, the above-mentioned step of: determining whether there is a vehicle with strong headlights among at least one moving vehicle based on image data includes: inputting the image data into a pre-trained first model, the first model being used to output at least one detection area, and the at least one detection area being used to represent the position information of at least one moving vehicle on the image data; inputting the image data of at least one detection area into a pre-trained second model, the second model being used to output a first confidence level of at least one detection area, and the first confidence level being used to represent the probability that there are strong headlights in at least one detection area; determining whether there is a vehicle with strong headlights among at least one moving vehicle based on the first confidence level of at least one detection area.

[0012] In this possible implementation, because the luminous intensity of the strong light is relatively consistent, generally higher than that of normal headlights, the license plate area can be exposed under relatively certain fill light conditions and exposure amounts to prevent overexposure or underexposure of the license plate. The light spot shape is similar when the strong light is imaged, and the brightness gradually decreases from the center to the surroundings, mostly in an elliptical shape. The relative position of the strong light on a moving vehicle is relatively fixed, mostly located in the center area below the front of the vehicle, adjacent to the license plate area, so the strong light vehicle can be detected through methods such as degree learning, which improves the feasibility of the solution.

[0013] In a possible implementation of the first aspect, the method also includes: inputting image data of at least one detection area into a pre-trained third model, the third model being used to output a license plate area in at least one detection area and a second confidence level of the license plate area, the second confidence level being used to indicate the probability that the license plate area is the license plate position of at least one moving vehicle; obtaining brightness information of the license plate area, and determining a third confidence level of at least one detection area based on the first confidence level and brightness information of the license plate area, the third confidence level being used to indicate the probability that a strong light exists in at least one detection area; the above-mentioned step of determining whether there is a vehicle with a strong light in at least one moving vehicle based on the first confidence level of at least one detection area includes: determining whether there is a vehicle with a strong light in at least one moving vehicle based on the third confidence level of at least one detection area.

[0014] In this possible implementation, license plate detection can also be combined to improve the detection accuracy of vehicles with strong headlights. Strong headlights will affect the brightness and clarity of the license plate area. Vehicles with strong headlights can also be understood as vehicles with unclear license plate areas. The higher the brightness represented by the brightness information of the license plate area, the lower the clarity of the license plate area, and the higher the probability that a strong headlight is installed near the license plate area.

[0015] In a possible implementation of the first aspect, the above-mentioned step: obtaining image data includes: obtaining video data, and the video data includes multiple frames of image data; the above-mentioned step: determining whether there is a vehicle with strong headlights among at least one moving vehicle based on the image data includes: determining whether there is a vehicle with strong headlights among at least one moving vehicle based on multiple frames of image data.

[0016] In this possible implementation, whether there is a vehicle with strong headlights can be determined based on multiple frames of image data, further improving the detection accuracy of vehicles with strong headlights.

[0017] In a possible implementation of the first aspect, the method also includes: generating a second image of a non-high-spot light vehicle based on a second shooting parameter and a second fill light parameter, the non-high-spot light vehicle being a vehicle other than a high-spot light vehicle among at least one moving vehicle, the second shooting parameter being a default shooting parameter, and the second fill light parameter being a default fill light parameter.

[0018] In this possible implementation, based on the detection results of vehicles detecting vehicles with strong headlights, when a vehicle with strong headlights turned on passes by, more appropriate fill light can be provided for the vehicle's area of ​​interest, and combined with the scene adaptation capability, more appropriate exposure can be provided for the captured image to ensure clear imaging and accurate intelligent recognition. When a moving vehicle passes by and the strong headlights are not detected to be turned on, lower intensity fill light and higher exposure are used to ensure less light pollution in most passing scenarios.

[0019] In a possible implementation manner of the first aspect, the method also includes: determining a first lane where a vehicle with strong headlights is located and a second lane where a vehicle without strong headlights is located; the above-mentioned step of: generating a first image of the vehicle with strong headlights based on the first shooting parameters and the first fill light parameters includes: adjusting the shooting parameters and fill light parameters used to shoot the first lane to the first shooting parameters and the first fill light parameters, and generating the first image; the above-mentioned step of: generating a second image of the vehicle without strong headlights based on the second shooting parameters and the second fill light parameters includes: adjusting the shooting parameters and fill light parameters used to shoot the second lane to the second shooting parameters and the second fill light parameters, and generating the second image.

[0020] In this possible implementation, different shooting parameters and fill light parameters are used for different lanes, which further ensures that the light pollution caused by the fill light is not increased while a clear snapshot of the vehicle with strong headlights can be obtained.

[0021] In a possible implementation manner of the first aspect, the adjustment priority of the first shooting parameter and the first fill-light parameter is higher than that of the second shooting parameter and the second fill-light parameter.

[0022] In this possible implementation, when different shooting parameters and fill-in light parameters are used for different lanes and there are both vehicles with strong headlights and vehicles without strong headlights, priority is given to ensuring the first shooting parameters and the first fill-in light parameters of the vehicle with strong headlights to ensure that a clear snapshot of the vehicle with strong headlights can be obtained.

[0023] In a possible implementation manner of the first aspect, the method further includes: generating a strong light mark of the first image; and performing image processing on the first image based on the strong light mark, the first shooting parameter, and the first fill light parameter.

[0024] In this possible implementation method, since the module used for image processing cannot determine whether the image to be processed is a snapshot of a vehicle with strong headlights, it can be determined based on the strong headlight mark, and more appropriate image processing can be performed on the snapshot of the vehicle with strong headlights, thereby further improving the imaging quality of the snapshot of the vehicle with strong headlights.

[0025] In a possible implementation of the first aspect, the above-mentioned step of: performing image processing on the first image based on the strong light mark, the first shooting parameter and the first fill light parameter includes: determining a first strong light suppression level of the first image based on the strong light mark, the first shooting parameter and the first fill light parameter; adjusting the strong light suppression level of the first image to the first strong light suppression level.

[0026] In this possible implementation method, the strong light suppression level of the captured image can be controlled specifically according to the shooting parameters and fill light parameters. Increasing the strong light suppression level can reduce the diffusion area and degree of the strong light halo, and reduce the impact of the strong light halo on the areas of interest such as license plates and windows, thereby ensuring that the brightness of the areas of interest is moderate and the clarity is higher.

[0027] In a possible implementation manner of the first aspect, the first shooting parameter includes an exposure parameter and a white balance parameter, and the first fill light parameter includes fill light brightness and fill light color temperature.

[0028] In this possible implementation, the first shooting parameter and the second shooting parameter are specifically different exposure parameters and white balance parameters, and the first fill light parameter and the second fill light parameter are specifically different fill light brightness and fill light color temperature, which improves the feasibility of the solution.

[0029] In a second aspect of the present application, a computing device is provided for executing the method in the first aspect or any possible implementation of the first aspect. Specifically, the computing device includes a module or unit for executing the method in the first aspect or any possible implementation of the first aspect, such as: an acquisition unit, a first determination unit, a second determination unit, a third determination unit, a fourth determination unit, a first generation unit, a second generation unit, and a third generation unit.

[0030] The third aspect of the present application provides a computing device, which includes a processor, a memory, and a computer-readable storage medium storing a computer program; the processor is coupled to the computer-readable storage medium, and the computer executes instructions running on the processor. When the computer executes the instructions, the processor executes the method of the first aspect or any possible implementation of the first aspect. Optionally, the computing device may also include an input / output (I / O) interface, and the computer-readable storage medium storing the computer program may be a memory.

[0031] The fourth aspect of the present application provides a computer-readable storage medium storing one or more computer-executable instructions. When the computer-executable instructions are executed by a processor, the processor executes a method as described in the first aspect or any possible implementation of the first aspect.

[0032] The fifth aspect of the present application provides a computer program product storing one or more computer-executable instructions. When the computer-executable instructions are executed by a processor, the processor executes a method as described in the first aspect or any possible implementation of the first aspect.

[0033] In a sixth aspect of the present application, a chip system is provided, the chip system comprising at least one processor and an interface, the interface being used to receive data and / or signals, and at least one processor being used to support a computer device to implement the functions involved in the above-mentioned first aspect or any possible implementation of the first aspect. In one possible design, the chip system may also include a memory, the memory being used to store program instructions and data necessary for the computer device. The chip system may be composed of a chip, or may include a chip and other discrete devices.

[0034] In an embodiment of the present application, image data is acquired, which includes images of moving vehicles. Then, based on the image data, it is determined whether there is a vehicle with strong light lights whose headlight brightness is greater than a first threshold value among the moving vehicles. If so, the shooting parameters and fill light parameters of the vehicle with strong light lights are determined, and an image of the vehicle with strong light lights is generated based on the shooting parameters and fill light parameters, thereby achieving clear imaging of the vehicle with strong light lights. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 A schematic diagram of a vehicle with strong headlights;

[0036] Figure 2 Schematic diagram of the scene captured for traffic;

[0037] Figure 3 A schematic diagram of an embodiment of a vehicle photographing method provided in an embodiment of the present application;

[0038] Figure 4-Figure 7 A schematic diagram of an embodiment of a camera system provided in an embodiment of the present application;

[0039] Figure 8 A schematic diagram of an embodiment of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0040] The following describes the embodiments of the present application in conjunction with the accompanying drawings. Obviously, the described embodiments are only embodiments of a part of the present application, rather than all embodiments. It is known to those skilled in the art that with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0041] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0042] The word “exemplary” is used exclusively herein to mean “serving as an example, example, or illustration.” Any embodiment described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.

[0043] In addition, in order to better illustrate the present application, numerous specific details are given in the following specific embodiments. It should be understood by those skilled in the art that the present application can also be implemented without certain specific details. In some examples, methods, means, components and circuits well known to those skilled in the art are not described in detail in order to highlight the subject matter of the present application.

[0044] The following is an example of the application scenario involved in the embodiments of the present application.

[0045] With the rapid development of camera technology, the application of cameras is becoming more and more extensive. For example, in traffic capture camera systems, it is necessary to realize video-based situation capture and image-based license plate capture. Traffic capture camera systems generally include cameras and fill lights, and the camera's shooting parameters include shutter, gain, and white balance.

[0046] Specifically, the shutter controls the length of time the image sensor is sensitive to light. The larger the shutter, the longer the sensitivity time, the more sufficient the exposure, the higher the image brightness, and the easier it is to produce motion ghosting. The gain controls the amplification factor of the image sensor's analog-to-digital conversion. The larger the gain, the higher the image brightness and the greater the image noise. White balance is used to eliminate color cast problems in camera imaging that do not conform to the real scene. The purpose is to make white objects seen by the human eye appear the same white in the image.

[0047] Fill-in lights commonly use light-emitting diode (LED) lamps and xenon lamps. LED lamps are fill-in lights with light sources provided by light-emitting diodes. Light-emitting diodes are solid-state semiconductor devices that can convert electrical energy into visible light. They can directly convert electricity into light. Due to their obvious advantages in terms of service life, energy saving and environmental protection, they are currently widely used. In the security industry, LED lamps are often used to fill in the light for cameras. LEDs can work in strobe mode or in constant light mode. The power of LEDs can also be flexibly controlled to provide different fill-in intensities. Instantly providing high-current and high-intensity fill-in is also called LED frequency exposure. Therefore, there are some sayings of LED constant light, LED strobe and LED flash. Xenon lamps can also be called gas flash lamps. They are a new type of headlight containing xenon gas, also known as high-intensity discharge gas lamps. By filling a variety of chemical gases in an anti-ultraviolet crystal quartz glass tube, using high voltage to separate them, a light source is generated between the two poles of the power supply, providing stable and high-intensity lighting. They are widely used in car headlights, photography fill-in, and snapshot fill-in.

[0048] like Figure 1 As shown in the figure, in scenes such as highways and expressways, especially at night, it is common to see some vehicles equipped with very bright strong lights near the license plate. The strong lights are not factory configurations of the vehicles, and they have high luminous intensity and dazzling light spots. Because the strong lights are installed close to the license plate, the light spots (as shown in the dotted box in the figure) are easy to spread to the license plate area, resulting in the general traffic capture camera system being unable to clearly image the license plate and unable to manage traffic behavior based on the license plate information.

[0049] Based on this, the embodiment of the present application provides a vehicle shooting method for achieving clear imaging of a vehicle with a strong headlight. The embodiment of the present application also provides a corresponding camera system, a computing device, and a computer-readable storage medium, which are described in detail below.

[0050] like Figure 2 As shown, the vehicle shooting method provided in the embodiment of the present application can be applied to the intelligent transportation system checkpoint capture scene. Among them, the camera system includes a camera and a fill light, both of which are installed on a pole. One camera covers multiple lanes, taking a two-lane checkpoint as an example. Generally speaking, each lane needs to be equipped with a fill light to ensure that the fill light spot covers the vehicle body area, and the camera is connected to multiple fill lights through a signal line. When a vehicle in a lane is driving towards the camera, it starts to capture when it passes the detection line, and the fill light of the lane where it is located flashes synchronously. The camera generates a snapshot that can clearly reflect the information of the vehicle, which is convenient for the management of road traffic behavior.

[0051] The moving vehicle in the present application is a vehicle in a broad sense, which can be a means of transportation (such as commercial vehicles, passenger cars, motorcycles, flying cars, trains, etc.), industrial vehicles (such as forklifts, trailers, tractors, etc.), engineering vehicles (such as excavators, bulldozers, cranes, etc.), agricultural equipment (such as mowers, harvesters, etc.), amusement equipment, toy vehicles, etc. The embodiments of the present application do not specifically limit the type of vehicle.

[0052] The vehicle shooting method provided in the embodiment of the present application is described below in combination with the above application scenarios.

[0053] like Figure 3 As shown, an embodiment of the present application provides a vehicle photographing method, and an embodiment of the method includes:

[0054] 301. Obtain image data.

[0055] The vehicle shooting method provided in the embodiment of the present application is applied to Figure 2 The camera system shown in the figure shoots at a road checkpoint to obtain image data, and the image data includes an image of at least one moving vehicle, that is, in the scene of the road checkpoint, if there is at least one moving vehicle, the camera system shoots and obtains image data. Alternatively, the camera system continuously shoots, and only when there is at least one moving vehicle, that is, when a moving vehicle passes by, does it obtain or save image data for subsequent processing.

[0056] Optionally, the camera system may also acquire video data within a period of time, where the video data includes multiple frames of image data, that is, the video data may be decomposed into multiple frames of image data, and the multiple frames of image data may be used for subsequent processing.

[0057] 302. Determine whether there is a vehicle with high beam lights among at least one moving vehicle based on the image data.

[0058] After the camera system acquires the image data, it determines whether there is a vehicle with high-spot lights among at least one moving vehicle based on the image data, wherein the brightness of the lights of the vehicle with high-spot lights is greater than a first threshold, that is, the vehicle with high-spot lights is installed and turned on, the high-spot lights are not a factory configuration of the vehicle, the luminous intensity is high, the brightness is greater than the first threshold, and the light spot is dazzling.

[0059] Because the luminous intensity of the strong light is relatively consistent, generally higher than that of normal headlights, the license plate area can be exposed under relatively certain fill light conditions and exposure amounts to prevent overexposure or underexposure of the license plate. The light spot shape when the strong light is imaged is similar, and the brightness gradually decreases from the center to the surroundings, mostly in an elliptical shape. The relative position of the strong light on a moving vehicle is relatively fixed, mostly located in the center area below the front of the vehicle, adjacent to the license plate area, so the strong light can be detected through pattern recognition, deep learning in the field of artificial intelligence, and other methods, and the hardware cost is reduced.

[0060] Specifically, when determining whether there is a vehicle with strong headlights, the image data is input into a pre-trained first model, and the first model is used to output at least one detection area, and the at least one detection area is used to represent the position information of at least one moving vehicle on the image data, that is, the first model is used to output the vehicle area corresponding to all vehicles in the image data, which is used to indicate the positions of all vehicles in the picture of the image data, and each detection area represents a moving vehicle.

[0061] After determining the positions of all moving vehicles in the picture corresponding to the image data, the image data of each detection area is intercepted, and the image data of at least one detection area is input into a pre-trained second model. The second model is used to output a first confidence level of at least one detection area, wherein the first confidence level is used to indicate the probability that there are strong headlights in at least one detection area. Then, based on the first confidence level of at least one detection area, it is determined whether there is a vehicle with strong headlights among at least one moving vehicle.

[0062] Specifically, the image data corresponding to each detection area is further partitioned, each detection area is divided into multiple detection sub-areas, and then the multiple detection sub-areas are input into the second model, and the second model outputs the first confidence of each detection sub-area. When there are multiple detection sub-areas in a detection area whose first confidence is higher than the first threshold value, it is determined that the first confidence of the detection area is higher than the first threshold value, and the vehicle corresponding to the detection area is a high-beam vehicle, and there is a high-beam vehicle among at least one moving vehicle.

[0063] Optionally, the image data of at least one detection area may be input into a pre-trained third model, and the third model is used to output the license plate area and the second confidence of the license plate area in the at least one detection area, and the second confidence is used to indicate the probability that the license plate area is the license plate position of at least one moving vehicle. Then, the brightness information of the license plate area is obtained, and the third confidence of at least one detection area is determined based on the second confidence and brightness information of the license plate area, and the third confidence is used to indicate the probability that there is a strong light in the at least one detection area. Finally, based on the first confidence and the third confidence of the at least one detection area, it is determined whether there is a strong light vehicle in the at least one moving vehicle.

[0064] Specifically, a strong light will affect the brightness and clarity of the license plate area, and a vehicle with a strong light can also be understood as a vehicle with an unclear license plate area. The third model can be understood as a license plate detection model, which is used to determine the license plate position in the vehicle. The image data of the detection area or detection sub-area is input into the third model, and the third model determines the second confidence of each detection area or detection sub-area. When the second confidence is higher than the second threshold value, the detection area or detection sub-area is determined to be the license plate area of ​​the moving vehicle, and then the brightness information of the license plate area is determined. The higher the brightness represented by the brightness information, the lower the clarity of the license plate area, and the higher the probability that a strong light is installed in the license plate area, the higher the first confidence. Therefore, the corresponding third confidence can be generated in combination with the first confidence and the brightness information, and whether the moving vehicle is a strong light vehicle is determined based on the third confidence.

[0065] Exemplarily, there are moving vehicles A and moving vehicles B in the image data, the first confidence of moving vehicle A (detection area A) is 0.5, detection area A is divided into multiple detection sub-areas, and the license plate area and the brightness information of the license plate area are determined to be 1 (the range of brightness information is 0 to 1, the higher the brightness, the higher the brightness), so the first confidence is corrected based on the brightness information, and the final third confidence is determined to be 0.9. The first confidence of moving vehicle B (detection area B) is 0.9, detection area A is divided into multiple detection sub-areas, and the license plate area and the brightness information of the license plate area are determined to be 0.1, and the final third confidence is determined to be 0.5. When the third confidence is higher than 0.8, it is determined that the moving vehicle corresponding to the detection area is a high-beam vehicle, so there is a high-beam vehicle between moving vehicle A and moving vehicle B, and the high-beam vehicle is moving vehicle A.

[0066] Optionally, when the acquired image data is multi-frame image data, each frame of image data can be input into the first model, the second model or the third model, and then the corresponding first confidence level or third confidence level can be obtained. The results of the multi-frame image data can be combined to comprehensively determine whether there is a vehicle with strong headlights among at least one moving vehicle and the position of the vehicle with strong headlights. The embodiments of the present application will not be described in detail here.

[0067] Optionally, image recognition can be performed directly on the image data to automatically identify vehicles with strong headlights. Optional image recognition methods include, but are not limited to, template matching, image feature recognition based on scale-invariant feature transform (SIFT) operator, image feature recognition based on speeded up robust features (SURF) operator, etc.

[0068] It should be understood that the above-mentioned first model, second model and third model are all deep neural network models, which are pre-trained with a large amount of sample data. For example, when the first model is trained, the sample data includes corresponding image data and detection areas, wherein the image data is used as input and at least one detection area is used as output for training. After the image data is input, the trained first model can output at least one detection area corresponding to the image data, that is, the position of the moving vehicle. The second model is similar to the third model, and the embodiments of the present application will not be repeated here.

[0069] 303. Determine a first shooting parameter and a first fill light parameter of the vehicle with a strong headlight.

[0070] 304. Generate a first image of the vehicle with a strong light based on the first shooting parameter and the first fill light parameter.

[0071] If there is a vehicle with a strong headlight among at least one of the moving vehicles, the first shooting parameter and the first fill light parameter of the vehicle with a strong headlight, and the second shooting parameter and the second fill light parameter of the vehicle without a strong headlight are determined. The vehicle without a strong headlight is a vehicle other than the vehicle with a strong headlight among the at least one moving vehicle, that is, the vehicle without a strong headlight is a normal vehicle without a strong headlight installed. The second shooting parameter is a default shooting parameter, and the second fill light parameter is a default fill light parameter, that is, the camera system uses the second shooting parameter and the second fill light parameter for snapshot by default, and only when there is a vehicle with a strong headlight, the first shooting parameter and the first fill light parameter are used for snapshot.

[0072] After determining the first shooting parameter and the first fill light parameter, a snapshot is taken based on the first shooting parameter and the first fill light parameter to generate a snapshot of a vehicle with a strong headlight, i.e., a first image, or a snapshot is taken based on the second shooting parameter and the second fill light parameter to generate a snapshot of a vehicle without a strong headlight, i.e., a second image. The first shooting parameter and the second shooting parameter include an exposure parameter and a white balance parameter, and the first fill light parameter and the second fill light parameter include the brightness of the fill light and the color temperature of the fill light.

[0073] Exemplarily, the fill light used to capture vehicles with strong headlights is a xenon lamp, the color temperature range of which is 5000K to 5500K, and according to the mapping relationship between the xenon lamp and the white area detection range in the configuration parameters, the white balance parameters are updated in time, that is, the white area detection range is updated in time to the white area detection range corresponding to the color temperature range of 5000K to 5500K. The fill light used to capture vehicles with non-strong headlights is an LED lamp, the color temperature range of which is 3500K to 4000K, and according to the mapping relationship between the LED lamp and the white area detection range in the configuration parameters, the white balance parameters are updated in time, that is, the white area detection range is updated in time to the white area detection range corresponding to the color temperature range of 3500K to 4000K.

[0074] Optionally, when generating the first image or the second image, the first lane where a vehicle with high headlights is located and the second lane where a vehicle without high headlights is located can also be determined, that is, in the image data, if there is a vehicle with high headlights on the lane, then the lane is the first lane; if there is no vehicle with high headlights on the lane, that is, there are only vehicles without high headlights on the lane, then the lane is the second lane, and different shooting parameters and fill light parameters are used for the first lane and the second lane. The shooting parameters and fill light parameters used to shoot the first lane are adjusted to the first shooting parameters and the first fill light parameters, and the first image is generated; the shooting parameters and fill light parameters used to shoot the second lane are adjusted to the second shooting parameters and the second fill light parameters, and the second image is generated.

[0075] Optionally, the adjustment priority of the first shooting parameter and the first fill light parameter is higher than that of the second shooting parameter and the second fill light parameter. For example, there are both vehicles with strong headlights and vehicles without strong headlights in lane A. In this case, the shooting parameter of the lane is adjusted to the first shooting parameter, and the fill light parameter is adjusted to the first fill light parameter.

[0076] Optionally, while generating the first image, a strong light mark of the first image may be further generated, and image processing may be performed on the first image based on the strong light mark, the first shooting parameter, and the first fill light parameter. Optionally, a first strong light suppression level of the first image may be determined based on the strong light mark, the first shooting parameter, and the first fill light parameter, and the strong light suppression level of the first image may be adjusted to the first strong light suppression level.

[0077] In the embodiment of the present application, based on the detection results of the vehicle detecting the vehicle with the strong light on, when the vehicle with the strong light on passes by, the vehicle's area of ​​interest (license plate area) can be more appropriately filled with light, and combined with the scene adaptation capability, the captured image (first image) can be more appropriately exposed to ensure clear imaging and accurate intelligent recognition; and when a moving vehicle passes by and the strong light is not detected to be turned on, a lower intensity fill light and a higher exposure are used to ensure less light pollution in most passing scenarios.

[0078] The vehicle photographing method is further illustrated below with reference to the camera system provided in the embodiment of the present application.

[0079] like Figure 4As shown, the embodiment of the present application also provides a camera system, the hardware structure of which includes a fill light, a lens, an image sensor, a field programmable gate array (FPGA) and a system on chip (SOC). The fill light provides snap fill light, and after being collected by the lens and the image sensor, the raw data is obtained, and then the image is processed and controlled by the SOC, and the fill light is controlled by the FPGA, such as controlling its fill light parameters, and the shooting parameters of the lens and the image sensor, such as controlling the exposure, to obtain a high-quality color image.

[0080] like Figure 5 As shown, the software structure of the camera system includes a shooting module, a detection module, a main control module, a capture module, a fill light module, a scene adaptation module, a capture image generation module and a capture image processing module.

[0081] Specifically, the shooting module includes a lens, an image sensor and an SOC, and generates images and video streams that can be used for vehicle detection and strong light detection, that is, it includes image data of at least one moving vehicle. The detection module is located in the SOC, detects the vehicle based on the image data, triggers the capture and preliminarily determines the existence area of ​​the strong light, that is, it is used to determine whether there is a strong light vehicle in at least one moving vehicle based on the image data. The main control module is located in the SOC, and takes the detection result of the detection module as input to control the scene adaptation module and the capture module. The capture module is located in the FPGA, and is informed of the fill light parameters by the main control module to control the fill light module, and is informed of the shooting parameters by the automatic exposure control module to control the capture image generation module. The fill light module is located in the fill light, and is used to receive the light control command of the capture module, and is a logic circuit acting on the fill light. The scene adaptation module is located in the SOC, and is informed by the main control module whether there is a strong light vehicle, and includes an automatic exposure control module and an automatic white balance control module. The capture image generation module is located in the image sensor, and adjusts the exposure of the capture image by receiving the shooting parameters sent by the capture module. The snapshot image processing module is located in the SOC, receives the shooting parameters passed by the scene adaptation module for image processing of the snapshot image (first image), receives the original image passed by the snapshot image generation module, and determines whether a snapshot image has adopted different fill light parameters and shooting parameters according to whether there is a strong light mark. Combining the input information of the above two, linking the parameters of other image processing modules, more appropriate image processing is performed on snapshot images under different fill light parameters and shooting parameters.

[0082] like Figure 6As shown, the camera system can also be divided into an intelligent layer, a main control layer and an image processing layer. When the camera system provided in the embodiment of the present application is applied to the vehicle shooting method provided in the embodiment of the present application, the monitoring module is used to execute step 301 in the above method embodiment to obtain image data.

[0083] The detection module is used to execute step 302 in the above method embodiment, and determine whether there is a vehicle with strong headlights in at least one moving vehicle based on the image data. Vehicle detection can adopt a general vehicle detection method. Its input can be a single frame image or a video stream. Since strong headlights will affect the imaging of passing vehicles, it is necessary to add special training data to the training set of the first model to ensure the accuracy and recall rate of the detection of vehicles with strong headlights.

[0084] The main control module uses the detection results of the detection module as input to control the scene adaptation module and the capture module. For example, if a vehicle with strong headlights is detected, the capture module and the scene adaptation module are informed to switch to the light control strategy (such as fill light parameters) and scene adaptation strategy (such as shooting parameters) for vehicles with strong headlights turned on for only that lane; if a vehicle with strong headlights not turned on is detected passing through a lane, the capture module and the scene adaptation module are informed to switch to the light control strategy and scene adaptation strategy for vehicles with strong headlights not turned on for only that lane.

[0085] If vehicles pass through several lanes at the same time, or the interval between vehicles passing through a lane is short, resulting in the coexistence of vehicles with and without strong beam lights within a period of time, making it impossible to switch the light control strategy and scene adaptation strategy in time, the same light control strategy and scene adaptation strategy for vehicles with strong beam lights on will be used for several nearby vehicles to ensure that the license plate image of the vehicle with strong beam lights on is clear.

[0086] The scene adaptation module is informed by the main control module whether the strong light is turned on when the vehicle passes by, including the automatic exposure control module and the automatic white balance control module.

[0087] Among them, the automatic exposure control module cooperates with the lighting control strategy of the main control module to adaptively adjust the exposure control strategy. Ensure that the area of ​​interest (license plate area) of the moving vehicle is properly exposed and the image is clear to prevent overexposure or underexposure. The exposure parameters are passed to the snapshot image generation module through the capture module. When a moving vehicle with a strong light turns on passes by, the exposure module reduces the exposure. Considering the adjustment speed, it generally adjusts the exposure time and gain. If the luminous intensity of the fill light per unit time increases and the luminous time shortens, the exposure time is shortened synchronously and the gain is adjusted as needed. If the luminous intensity of the fill light per unit time increases but the luminous time remains unchanged, the gain is reduced synchronously and the exposure time is adjusted as needed.

[0088] The automatic white balance control module cooperates with the lighting control strategy of the main control module to adaptively adjust the white balance control strategy. Ensure that the white balance parameters of the fill light area are appropriate and there is no color cast during imaging. Pass the white balance parameters to the snapshot image processing module. For example, in the calibration stage of generating the first image, it is necessary to know in advance the type of fill light corresponding to the vehicle snapshot with and without the strong light turned on. If the fill light types are different, the fill light type and fill light color temperature range corresponding to the vehicle snapshot with the strong light turned on, and the fill light type and fill light color temperature range corresponding to the vehicle snapshot without the strong light turned on can be determined respectively. Calibrate the white balance parameters corresponding to the fill light color temperature range in the above two cases, such as the white area detection range in the white balance parameters. And store the mapping relationship between the above fill light type and the above white area detection range in the configuration parameters of the automatic white balance control module. When the fill light module switches to the corresponding fill light according to whether the pre-captured vehicle has a strong light or not, the white area detection range of the automatic white balance control module is immediately constrained to the white area detection range corresponding to the pre-switched fill light type in the above configuration parameters, so that the white area detection range in the white balance parameters is updated in time to prevent color cast of the picture caused by failure to update the white area detection range in time.

[0089] The capture module is informed of the light control strategy by the main control module to control the fill light module; the automatic exposure control module is informed of the exposure strategy to control the capture image generation module. When a vehicle with strong lights passes by, the capture module is informed of the light control strategy by the main control module to enable the fill light module to use a higher luminous intensity. The automatic exposure control module is informed of the exposure strategy and sends it to the capture image generation module. When a vehicle with strong lights passes by, the capture module is informed of the exposure strategy by the automatic exposure control module to enable the capture image generation module to use a lower exposure for exposure. The capture module needs to ensure that the light control strategy and exposure strategy take effect simultaneously in one capture. The capture module needs to ensure that during one capture process, the light flashes and exposure are performed simultaneously, that is, the exposure time includes the light flashing time.

[0090] Optionally, the capture module tags the captured image (first image) when it is generated, so as to inform the captured image processing module whether a captured image has adopted different light control strategies and exposure strategies according to the presence or absence of a strong light.

[0091] The capture module and the main control module can be carried by two different hardwares respectively, or carried by one hardware alone.

[0092] The fill light module receives the light control command from the capture module and is a logic circuit acting on the fill light. When a vehicle passes by with the high beam light turned on, the fill light group switches to a fill light with higher luminous intensity, such as switching from an LED lamp to a xenon lamp, compared to a vehicle passing by without the high beam light turned on. Or the luminous intensity of the fill light is improved, such as increasing the luminous intensity of the LED lamp, or increasing the luminous intensity of the xenon lamp. A higher luminous intensity can be understood as a higher luminous intensity per unit time, a shorter or unchanged luminous time, and even a lower total luminous amount.

[0093] The snapshot image generation module receives the exposure parameters sent by the snapshot module and acts on the photosensitive sensor to adjust the exposure of the snapshot image. For example, when a vehicle with a strong light passes by, the photosensitive sensor uses a lower exposure than when a vehicle without a strong light passes by.

[0094] The captured image processing module receives the shooting parameters transmitted by the scene adaptation module for processing the captured image. The shooting parameters include but are not limited to exposure parameters and white balance parameters.

[0095] Because the light control strategy is sent directly from the main control module to the capture module, and the exposure strategy is sent from the main control module to the capture module via the automatic exposure module. Therefore, the capture module needs to inform the capture image processing module whether the capture image adopts the light control strategy and exposure strategy for the strong light. Therefore, the capture module also needs to mark the original image with a label indicating whether there is a strong light (strong light mark). Receive the original image passed by the capture image generation module, and determine whether a certain capture image has adopted different light control strategies and exposure strategies based on the presence or absence of a strong light based on the label or mark of the strong light. Combining the input information of the above two, the parameters of other image processing modules are linked to perform more appropriate image processing on the capture images under different fill light parameters and shooting parameters.

[0096] For example, if it is a vehicle with strong headlights, the image processing module can increase the strong light suppression level. The increase in the strong light suppression level can reduce the diffusion area and degree of the strong light halo, reduce the impact of the strong light halo on areas of interest such as license plates and windows, and ensure that the brightness of the areas of interest is moderate and the clarity is higher.

[0097] By using the vehicle shooting method or camera system provided by the embodiment of the present application, when a vehicle with a strong light turns on passes by, the fill light intensity will be improved, which is manifested as switching to a fill light with higher luminous intensity, or the luminous intensity of the fill light itself is increased. And there is a relatively definite correspondence between whether the strong light is turned on and the fill light intensity when a moving vehicle passes by. In addition, when the fill light intensity increases due to the passage of a moving vehicle with a strong light turned on, it can be ensured that the picture of the snapshot is not overexposed, especially the area of ​​interest is not overexposed. When the fill light intensity decreases due to the passage of a moving vehicle without a strong light turned on, it can be ensured that the picture of the snapshot is not underexposed, especially the area of ​​interest is not underexposed, and when the color temperature of the fill light changes significantly, it can be ensured that the fill light spot area in the picture is not cast in color.

[0098] like Figure 7 As shown, in another embodiment of the camera system provided by the embodiment of the present application, the camera system 700 includes:

[0099] The acquisition unit 701 is used to acquire image data, where the image data includes at least one image of a moving vehicle. The acquisition unit 701 can execute step 301 in the above method embodiment.

[0100] The first determination unit 702 is used to determine whether there is a vehicle with strong headlights among at least one moving vehicle based on the image data, and the headlight brightness of the vehicle with strong headlights is greater than a first threshold; the first determination unit 702 can execute step 302 in the above method embodiment.

[0101] The second determining unit 703 is used to determine the first shooting parameter and the first fill light parameter of the high-spot light vehicle if there is a high-spot light vehicle among at least one of the moving vehicles; the second determining unit 703 can execute step 303 in the above method embodiment.

[0102] The first generating unit 704 is configured to generate a first image of the vehicle with a strong headlight based on the first shooting parameter and the first fill light parameter. The first generating unit 704 may execute step 304 in the above method embodiment.

[0103] Optionally, the first determination unit 702 is specifically used to input image data into a pre-trained first model, the first model is used to output at least one detection area, and the at least one detection area is used to represent the position information of at least one moving vehicle on the image data; the image data of at least one detection area is input into a pre-trained second model, the second model is used to output a first confidence level of at least one detection area, and the first confidence level is used to represent the probability that there are strong light lights in at least one detection area; based on the first confidence level of at least one detection area, it is determined whether there is a vehicle with strong light lights among at least one moving vehicle.

[0104] Optionally, the camera system 700 also includes a third determination unit 705, which is used to input image data of at least one detection area into a pre-trained third model, and the third model is used to output a license plate area in at least one detection area and a second confidence level of the license plate area, and the second confidence level is used to indicate the probability that the license plate area is the license plate position of at least one moving vehicle; obtain brightness information of the license plate area, and determine a third confidence level of at least one detection area based on the first confidence level and brightness information of the license plate area, and the third confidence level is used to indicate the probability that there is a strong light in at least one detection area; the first determination unit 702 is specifically used to determine whether there is a strong light vehicle among at least one moving vehicle based on the third confidence level of at least one detection area.

[0105] Optionally, the acquisition unit 701 is specifically used to acquire video data, and the video data includes multiple frames of image data; the first determination unit 702 is specifically used to determine whether there is a vehicle with strong headlights among at least one moving vehicle based on the multiple frames of image data.

[0106] Optionally, the camera system 700 also includes a second generation unit 706, which is used to generate a second image of a non-high-spot light vehicle based on second shooting parameters and second fill light parameters. The non-high-spot light vehicle is a vehicle other than the high-spot light vehicle among at least one moving vehicle. The second shooting parameters are default shooting parameters, and the second fill light parameters are default fill light parameters.

[0107] Optionally, the camera system 700 also includes a fourth determination unit 707, which is used to determine the first lane where the vehicle with high beam lights is located and the second lane where the vehicle without high beam lights is located; the first generation unit 704 is specifically used to adjust the shooting parameters and fill light parameters used to shoot the first lane to first shooting parameters and first fill light parameters, and generate a first image; the second generation unit 706 is specifically used to adjust the shooting parameters and fill light parameters used to shoot the second lane to second shooting parameters and second fill light parameters, and generate a second image.

[0108] Optionally, the adjustment priority of the first shooting parameter and the first fill-light parameter is higher than that of the second shooting parameter and the second fill-light parameter.

[0109] Optionally, the camera system 700 further includes a third generating unit 708, which is used to generate a strong light mark of the first image; and perform image processing on the first image based on the strong light mark, the first shooting parameter and the first fill light parameter.

[0110] Optionally, the third generating unit 708 is specifically configured to determine a first strong light suppression level of the first image based on the strong light mark, the first shooting parameter and the first fill light parameter; and adjust the strong light suppression level of the first image to the first strong light suppression level.

[0111] Optionally, the first shooting parameter includes an exposure parameter and a white balance parameter, and the first fill light parameter includes fill light brightness and fill light color temperature.

[0112] The camera system 700 provided in the embodiment of the present application may correspond to Figure 2 , Figure 5 or Figure 6 The camera system shown can also be understood by referring to the corresponding contents of the aforementioned license plate shooting method embodiment, which will not be repeated here.

[0113] like Figure 8 As shown, a possible logical structure diagram of a computing device 800 provided in an embodiment of the present application. The computing device 800 includes: a processor 801, a communication interface 802, a memory 803 and a bus 804. The processor 801 may include a CPU, or a CPU and at least one of a GPU, an NPU and other types of processors. The processor 801, the communication interface 802 and the memory 803 are interconnected via a bus 804. In an embodiment of the present application, the processor 801 is used to control and manage the actions of the computing device 800. For example, the processor 801 is used to execute the vehicle shooting method described in the above method part embodiment and / or other processes for the technology described herein. The communication interface 802 is used to support the computing device 800 to communicate. The memory 803 is used to store program code and data of the computing device 800.

[0114] Among them, the processor 801 can be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit, a field programmable gate array or other programmable logic device, a transistor logic device, a hardware component or any combination thereof. It can implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the contents disclosed in this application. The processor can also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a digital signal processor and a microprocessor, and so on. The bus 804 can be a peripheral component interconnect standard (Peripheral Component Interconnect, PCI) bus or an extended industry standard architecture (Extended Industry Standard Architecture, EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 8 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0115] In another embodiment of the present application, a computer-readable storage medium is provided, in which computer-executable instructions are stored. When at least one processor of a device executes the computer-executable instructions, the device executes the vehicle shooting method described in some embodiments of the above method.

[0116] In another embodiment of the present application, a computer program product is also provided, which includes computer execution instructions, which are stored in a computer-readable storage medium; at least one processor of the device can read the computer execution instructions from the computer-readable storage medium, and at least one processor executes the computer execution instructions so that the device executes the vehicle shooting method described in the above method embodiment.

[0117] In another embodiment of the present application, a chip system is also provided, the chip system includes at least one processor and an interface, the interface is used to receive data and / or signals, and the at least one processor is used to support the vehicle shooting method described in the above method part embodiment. In one possible design, the chip system may also include a memory, the memory is used to store the necessary program instructions and data of the computer device. The chip system can be composed of chips, and may also include chips and other discrete devices.

[0118] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the embodiments of the present application.

[0119] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0120] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0121] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0122] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0123] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, read-only memory), random access memory (RAM, random access memory), disk or optical disk and other media that can store program code.

Claims

1. A vehicle photographing method, characterized in that: include: Acquiring image data, the image data comprising at least one image of a moving vehicle; Determine whether there is a vehicle with strong headlights among the at least one traveling vehicle based on the image data, the headlight brightness of the vehicle with strong headlights being greater than a first threshold; If there is a high-beam light vehicle among the at least one moving vehicle, determining a first shooting parameter and a first fill light parameter of the high-beam light vehicle; A first image of the high-beam vehicle is generated based on the first shooting parameter and the first fill-light parameter.

2. The method according to claim 1, characterized in that The determining, based on the image data, whether there is a vehicle with strong headlights among the at least one moving vehicle comprises: Inputting the image data into a pre-trained first model, wherein the first model is used to output at least one detection area, wherein the at least one detection area is used to represent the position information of the at least one moving vehicle on the image data; Inputting the image data of the at least one detection area into a pre-trained second model, wherein the second model is used to output a first confidence level of the at least one detection area, wherein the first confidence level is used to indicate the probability that a strong light exists in the at least one detection area; Based on the first confidence level of the at least one detection area, it is determined whether there is a high-beam vehicle in the at least one traveling vehicle.

3. The method according to claim 2, characterized in that The method further comprises: Inputting the image data of the at least one detection area into a pre-trained third model, the third model is used to output a license plate area in the at least one detection area and a second confidence of the license plate area, the second confidence is used to represent the probability that the license plate area is the license plate position of the at least one moving vehicle; Acquire brightness information of the license plate area, and determine a third confidence of the at least one detection area based on the first confidence and brightness information of the license plate area, wherein the third confidence is used to indicate a probability that a strong light exists in the at least one detection area; Determining whether there is a high-beam vehicle among the at least one moving vehicle based on the first confidence level of the at least one detection area comprises: It is determined whether there is a high-beam vehicle among the at least one traveling vehicle based on a third confidence level of the at least one detection area.

4. The method according to any one of claims 1 to 3, characterized in that The acquiring of image data comprises: Acquire video data, wherein the video data includes a plurality of frames of image data; The determining, based on the image data, whether there is a vehicle with strong headlights among the at least one moving vehicle comprises: It is determined based on the multiple frames of image data whether there is a vehicle with high beam lights among the at least one moving vehicle.

5. The method according to any one of claims 1 to 4, characterized in that The method further comprises: A second image of a non-high-spot light vehicle is generated based on second shooting parameters and second fill light parameters, wherein the non-high-spot light vehicle is a vehicle other than the high-spot light vehicle among the at least one moving vehicle, the second shooting parameters are default shooting parameters, and the second fill light parameters are default fill light parameters.

6. The method according to claim 5, characterized in that The method further comprises: Determine a first lane where the vehicle with the high beam headlights is located and a second lane where the vehicle without the high beam headlights is located; The generating the first image of the vehicle with the strong light based on the first shooting parameter and the first fill light parameter comprises: adjusting shooting parameters and fill-in light parameters used for shooting the first lane to the first shooting parameters and the first fill-in light parameters, and generating the first image; Generating a second image of the vehicle without strong headlights based on the second shooting parameter and the second fill light parameter comprises: The shooting parameters and fill-in light parameters used for shooting the second lane are adjusted to the second shooting parameters and the second fill-in light parameters, and the second image is generated.

7. The method according to claim 6, characterized in that The adjustment priority of the first shooting parameter and the first fill-light parameter is higher than that of the second shooting parameter and the second fill-light parameter.

8. The method according to any one of claims 1 to 7, characterized in that The method further comprises: generating a strong light mark of the first image; Image processing is performed on the first image based on the high-intensity light mark, the first shooting parameter, and the first fill-light parameter.

9. The method according to claim 8, characterized in that The performing image processing on the first image based on the strong light mark, the first shooting parameter and the first fill light parameter comprises: Determine a first strong light suppression level of the first image based on the strong light mark, the first shooting parameter, and the first fill light parameter; The strong light suppression level of the first image is adjusted to the first strong light suppression level.

10. The method according to any one of claims 1 to 9, characterized in that The first shooting parameter includes an exposure parameter and a white balance parameter, and the first fill light parameter includes fill light brightness and fill light color temperature.

11. A camera system, characterized in that: include: an acquisition unit, configured to acquire image data, wherein the image data includes an image of at least one moving vehicle; A first determining unit, configured to determine whether there is a vehicle with strong headlights among the at least one traveling vehicle based on the image data, the headlight brightness of the vehicle with strong headlights being greater than a first threshold; A second determining unit is configured to determine a first shooting parameter and a first fill light parameter of a high-beam light vehicle if there is a high-beam light vehicle among the at least one moving vehicle; The first generating unit is used to generate a first image of the high-beam vehicle based on the first shooting parameter and the first fill-light parameter.

12. The system according to claim 11, characterized in that The first determination unit is specifically used to input the image data into a pre-trained first model, the first model is used to output at least one detection area, and the at least one detection area is used to represent the position information of the at least one moving vehicle on the image data; The image data of the at least one detection area is input into a pre-trained second model, and the second model is used to output a first confidence level of the at least one detection area, and the first confidence level is used to indicate the probability that a high-intensity light exists in the at least one detection area; based on the first confidence level of the at least one detection area, it is determined whether there is a high-intensity light vehicle among the at least one moving vehicle.

13. The system according to claim 12, characterized in that The system further comprises: a third determination unit, configured to input the image data of the at least one detection area into a pre-trained third model, wherein the third model is configured to output a license plate area in the at least one detection area and a second confidence level of the license plate area, wherein the second confidence level is used to indicate a probability that the license plate area is the license plate position of the at least one moving vehicle; obtain brightness information of the license plate area, and determine a third confidence level of the at least one detection area based on the first confidence level and the brightness information of the license plate area, wherein the third confidence level is used to indicate a probability that a strong light exists in the at least one detection area; The first determination unit is specifically configured to determine whether there is a high-beam vehicle among the at least one moving vehicle based on a third confidence level of the at least one detection area.

14. The system according to any one of claims 11 to 13, characterized in that: The acquisition unit is specifically used to acquire video data, wherein the video data includes multiple frames of image data; The first determination unit is specifically configured to determine whether there is a vehicle with strong headlights among the at least one moving vehicle based on multiple frames of image data.

15. The system according to any one of claims 11 to 14, characterized in that The system further comprises: The second generating unit is used to generate a second image of a non-high-spot light vehicle based on second shooting parameters and second fill light parameters, wherein the non-high-spot light vehicle is a vehicle other than the high-spot light vehicle among the at least one moving vehicle, the second shooting parameters are default shooting parameters, and the second fill light parameters are default fill light parameters.

16. The system according to claim 15, characterized in that The system further comprises: a fourth determining unit, configured to determine a first lane where the vehicle with the high beam headlights is located and a second lane where the vehicle without the high beam headlights is located; The first generating unit is specifically configured to adjust the shooting parameters and fill-in light parameters used for shooting the first lane to the first shooting parameters and the first fill-in light parameters, and generate the first image; The second generating unit is specifically configured to adjust the shooting parameters and fill-in light parameters used for shooting the second lane to the second shooting parameters and the second fill-in light parameters, and generate the second image.

17. The system according to claim 16, characterized in that The adjustment priority of the first shooting parameter and the first fill-light parameter is higher than that of the second shooting parameter and the second fill-light parameter.

18. The system according to any one of claims 11 to 17, characterized in that The system further comprises: The third generating unit is used to generate a strong light mark of the first image; and perform image processing on the first image based on the strong light mark, the first shooting parameter and the first fill light parameter.

19. The system according to claim 18, characterized in that The third generating unit is specifically used to determine a first strong light suppression level of the first image based on the strong light mark, the first shooting parameter and the first fill light parameter; and adjust the strong light suppression level of the first image to the first strong light suppression level.

20. The system according to any one of claims 11 to 19, characterized in that The first shooting parameter includes an exposure parameter and a white balance parameter, and the first fill light parameter includes fill light brightness and fill light color temperature.

21. A computing device, characterized in that The method comprises a processor on which a computer-executable instruction runs. When the computer-executable instruction is executed by the processor, the processor executes the method according to any one of claims 1 to 10.

22. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 10 is implemented.

Images