A low-light fill-in capture system for a moving vehicle and a control method thereof

Abstract

The present invention relates to a system for filling light and capturing a vehicle under low illumination and a control method thereof, wherein the system comprises a filling light component, a video acquisition component, a video analysis component, image synthesis, a main control component and a timing position. The present invention adopts a multi-segment directional linear array controllable luminous lighting device, sets a multi-state mode of different filling light band lighting according to the distance of the vehicle from the capture pole, and uses a video acquisition component to perform infrared, visible light and ultraviolet full-band perception imaging; the system classifies the vehicle model, driver and passenger height, etc. of the vehicle and calculates the speed through the video analysis component, and controls the filling light device through the main control component, and finally performs image filtering and compensation processing for the specific wavelength of the filling light, and completes the multi-information synthesis and superposition of the captured picture. The present invention solves the problem of energy-saving induction control of the filling light for capturing a vehicle and the impact of the filling light on the driver; and improves the vehicle model adaptability problem of the filling light capture illumination of the vehicle.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent road traffic and public safety monitoring, and in particular relates to a low-illuminance fill-in light capture system for a moving vehicle and a control method thereof. Background Art

[0002] With the development of society, the number of cars in cities has increased sharply, and related road traffic safety incidents have occurred frequently. It is an effective measure to use a capture system for moving vehicles on the road to trace and predict related public safety incidents. The capture system needs to record the panoramic features, license plates, facial features of front-seat drivers and passengers, etc. of passing vehicles. Given that light has a great impact on the camera, ordinary cameras are not ideal for capturing and distinguishing vehicles and license plates when the light is insufficient. Therefore, fill lights are usually added to the capture system to adjust the road lighting environment during capture.

[0003] In view of the high energy consumption of constant lighting, short-term flash is usually used for vehicle snapshot fill-in lighting. At night, the sudden high-brightness flash of the flash can easily cause great fluctuations and influences on the driver's vision and psychology, and may even cause traffic accidents when driving at high speeds. The fill-in intensity required for capturing vehicles is different. The fill-in lighting requirements for capturing information such as the appearance and outline of the vehicle model are relatively weak. When capturing vehicle exterior details such as license plates, logos, and faces, moderate fill-in lighting is required in conjunction with the imaging technology of multiple exposures and strong light suppression. When the snapshot image needs to clearly show the facial features of the front driver and passenger, the fill-in intensity required is the highest, and the fill-in light needs to penetrate the driver's window glass and explosion-proof film to effectively illuminate the front driver and passenger seats. At the same time, due to the different driver and passenger positions of different types of vehicles, efficient snapshot fill-in lighting is more difficult. Current related systems and devices have made some improvements to these problems, using long-term high-frequency flashing of LEDs, or gradually increasing the light intensity to reduce the stimulation to the driver's eyes, but the performance is not perfect. Some fill-in-light devices that automatically adjust the angle are too complicated in design, have slow response speed and poor reliability, and cannot solve the problem well. Summary of the invention

[0004] In order to overcome the deficiencies of the prior art, effectively alleviate the irritation of the light environment to the human eye and improve the effectiveness of capturing, the present invention provides a low-illuminance fill-light capture system for a moving vehicle and a control method thereof, which can capture the features of moving vehicles and drivers and passengers on the road under low-illuminance conditions.

[0005] The technical solution adopted by the present invention is:

[0006] A low-light fill-in light capture system for a moving vehicle, the system comprising a fill-in light component, a video acquisition component, a video analysis component, an image synthesis component, a main control component and a timing position component;

[0007] The output end of the video acquisition component is connected to the video analysis component and the image synthesis component respectively; the image synthesis component outputs to the main control component and the video analysis component respectively; the output end of the video analysis component and the time position component is connected to the main control component, and the main control component output controls the fill light component.

[0008] Preferably, the video acquisition component adopts a fixed focal length, the infrared band focal length f(ir)> the visible light back depth of field focal length f(f)> the visible light focal length f(m)> the visible light front depth of field focal length f(n), and the video acquisition component has the imaging capability of sensing the infrared band, the entire visible light band and the ultraviolet band.

[0009] Preferably, the fill light component is an arc-shaped multi-band directional line array controllable fill light device, which is arranged in an arc shape toward the direction of vehicle travel to form a segmented capture fill light area, and the vertical area of ​​each segment of light irradiating the target is not less than 3 meters * 5 meters, with the fill light pole as the center and the height H; the distance between the pole and the irradiated target corresponds to the different focal length segments of the video acquisition component, which is L = f(x), wherein f(x) refers to the focal length, and x refers to the range of each segmented focal length segment; the projection angle of each segment of the fill light illumination arc is θ = arc tan(L / H); the ultra-long-distance infrared light-emitting component is located at the upper part of the component, irradiating the ultra-long distance L ir Target: The long-distance visible light emitting component located in the upper middle part of the component illuminates the long-distance L f Target: The mid-range visible light emitting component located in the middle, illuminating the mid-range L m Target: The short-range visible light component located at the bottom of the component, irradiating the short distance L n Target: The focal length of the light-transmitting lens of the luminous component is also set in sections, focusing on each area to meet the imaging requirements of each section.

[0010] Preferably, the ultra-long distance is 30 meters ≤ L ir ≤50 meters, long distance is 20 meters ≤L f ≤30 meters, medium distance is 10 meters ≤L m ≤20 meters, short distance is L n ≤10 meters.

[0011] As a preferred option, ultra-long distance supplementary light wavelength λ ir Infrared band; long-distance fill light wavelength λ f The wavelength of the visible light component λ is within the green-yellow range of the visible light band, which is the part of the visible light band that the human eye is more sensitive to. Its luminous intensity is greater than the visible threshold of the human eye and less than the maximum value that the human eye can tolerate. m There are two control modes, the first wavelength λ m1 is the visible light band, yellow-red light range, λ m1 ≠λ f, the luminous intensity is greater than the long-distance fill light intensity and less than the maximum value that the human eye can tolerate; the second wavelength λ m2 It is a mixed wavelength of visible light and ultraviolet band, in the blue-violet light range, and its luminous intensity can meet the imaging requirements of personnel and characteristic targets in the vehicle driver's space; the wavelength of short-range fill light λ n =λ m2 The luminous effect should be able to meet the imaging requirements of personnel and characteristic targets in the vehicle driver and passenger space; the illumination value of each target section gradually increases from far to near.

[0012] Preferably, the video analysis component has the ability to perceive and analyze imaging light intensity, and can output two classification results of sufficient lighting and insufficient lighting. It also has the function of identifying external components including vehicle model, license plate, and vehicle logo based on infrared band imaging and visible light band imaging, and can output the height of the driver and passenger position from the ground in three categories according to infrared imaging: low, medium, and high.

[0013] Preferably, the video analysis component can mark the capture time of the first appearance of the infrared fill light imaging wavelength characteristics, the long-distance visible light fill light imaging wavelength characteristics, the medium-distance visible light fill light imaging wavelength characteristics, and the close-distance visible light fill light imaging characteristics according to the timing information, and calculate the vehicle's driving time T and driving speed V, V=S / T, S is one of the distance differences among the farthest position of the infrared fill light, the farthest position of the long-distance visible light, the farthest position of the medium-distance visible light, and the farthest position of the close-distance fill light.

[0014] The present invention also provides a control method for the low-light fill-light capture system for a moving vehicle, comprising the following steps:

[0015] S01: The video acquisition component captures the visible light image information of the vehicle target, and analyzes it through the video analysis component. When the output result is insufficient light, the main control component controls to turn on the ultra-long-distance infrared fill light;

[0016] S02: The video acquisition component captures the imaging information of the infrared characteristics of the vehicle target, records the imaging time t0, analyzes it through the video analysis component, classifies and outputs the vehicle type and the height of the driver and passenger position from the ground, and the main control component controls to turn on the long-distance fill light;

[0017] S03: The video acquisition component captures the visible light image information of the vehicle target under the long-distance fill light illumination characteristics, records the imaging time t1, calculates the vehicle speed V from time t0 to time t1 through the video analysis component, and measures the vehicle speed V at the medium distance L m The running time T in the range,

[0018] When the driver and passenger classification output is medium or high, within T time, the main control component controls to turn on and select the first type of medium-distance fill light, and the luminous brightness is gradually increased according to the vehicle speed ratio;

[0019] When the driver and passenger classification output is low, within T time, the main control component selects the first type of medium-distance fill light, and the luminous brightness is gradually increased according to the vehicle speed ratio, and then controls to turn on and select the second type of medium-distance fill light to achieve short-term lighting;

[0020] Under the first fill light illumination condition, after color filtering compensation is performed by the image synthesis component, the video analysis component obtains structured information including vehicle model and vehicle color;

[0021] Under the second fill light illumination condition, after color filtering compensation by the image synthesis component, the video analysis component obtains structural information including the position of the driver and passengers in the vehicle, and the faces and objects inside the vehicle;

[0022] S04: The video acquisition component captures the visible light image information of the vehicle target under the mid-range fill light illumination characteristics, records the imaging time t2, and when the driver and passenger classification output is medium or high, calculates the vehicle speed V' and the focal length ratio from time t0 to time t2, and measures the vehicle speed V' and the focal length ratio when entering the close distance L n At this time, the main control component briefly turns on the close-range fill light, and at the same time, after color filtering compensation by the image synthesis component, the video analysis component obtains structured information including the position of the vehicle driver and passengers, and the faces and objects inside.

[0023] S05: The video frame image with the highest definition is captured from each segmented video, and is superimposed and synthesized by an image synthesis component, and relevant structured information including time, vehicle speed, vehicle color, vehicle model, and license plate is annotated.

[0024] Preferably, the image component respectively generates the long-distance fill light illumination area image PIC-F, the medium-distance fill light illumination image PIC-M, and the short-distance fill light illumination image PIC-N according to λ f , m , n The corresponding wavelength is filtered and compensated, and then PIC-F, PIC-M, and PIC-N are superimposed and synthesized according to the vehicle profile information obtained by the video analysis component; when the synthesis and superposition are inconsistent, the vehicle model and color shall be based on PIC-F, the license plate and logo graphics shall be based on PIC-M, and the driver and passenger appearance images shall be based on PIC-N.

[0025] The present invention can be used in a moving vehicle capture system under low illumination, and the present invention has the following beneficial effects:

[0026] (1) It solves the problem of segmented fill light for capturing moving vehicles under low illumination conditions;

[0027] (2) The problem of arranging segmented fill light using lights of different characteristic bands is solved;

[0028] (3) Solved the color deviation problem of single-light band imaging;

[0029] (4) According to the driving conditions of the vehicle, the height of the driver and passengers is classified and identified and the speed is calculated to achieve efficient adjustment of the capture mode. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a structural schematic diagram of a low-illuminance fill-in light capture system for a moving vehicle according to the present invention;

[0031] Figure 2 This is a flow chart of a control method of a low-light fill-light capture system for a moving vehicle according to the present invention;

[0032] Figure 3 It is a schematic diagram of the projection effect of the arc-shaped multi-band directional fill light device of the present invention. DETAILED DESCRIPTION

[0033] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the present invention is further described below in conjunction with the accompanying drawings and embodiments, but the scope of protection of the present invention is not limited thereto.

[0034] Reference Figures 1 to 3 , a low-light fill-in light capture system for a moving vehicle, the system comprising a fill-in light component, a video acquisition component, a video analysis component, an image synthesis component, a main control component and a timing position (example: satellite positioning and timing module, GPS / Beidou component) component;

[0035] The output end of the video acquisition component is connected to the video analysis component and the image synthesis component respectively; the image synthesis component outputs to the main control component and the video analysis component respectively; the output end of the video analysis component and the time position component is connected to the main control component, and the main control component output controls the fill light component.

[0036] The video acquisition component adopts a fixed focal length, the infrared band focal length f(ir)> the visible light back depth of field focal length f(f)> the visible light focal length f(m)> the visible light front depth of field focal length f(n), and has the imaging capability of sensing the infrared band, the entire visible light band and the ultraviolet band (the number of collected samples: each imaging segment is higher than 6 frames).

[0037] The fill light component is an arc-shaped multi-band directional line array controllable fill light device, which is arranged in an arc shape toward the direction of vehicle travel to form a segmented capture fill light area. The vertical area of ​​the target illuminated by each segment of light is not less than 3 meters * 5 meters. The fill light pole is taken as the center (height H), and the distance between the pole and the illuminated target corresponds to different focal length segments of the video acquisition component, which is L=f(x), wherein f(x) refers to the focal length, and x refers to the range of each segmented focal length segment; the projection angle of each segment of the fill light illumination arc is θ=arctan(L / H); the ultra-long-distance infrared light-emitting component is located at the upper part of the component, illuminating the ultra-long-distance Lir Target (Example: 30m ≤ L ir ≤50 meters); the long-distance visible light emitting component located in the upper middle part of the component illuminates the long-distance L f Target (Example: 20m ≤ L f ≤30 meters); the medium-distance visible light emitting component located in the middle, irradiating the long-distance L m Target (Example: 10m ≤ L m ≤20 meters); the short-range visible light component located at the bottom of the component illuminates the long-range L n Target (Example: L n ≤10 meters); the focal length of the light-transmitting lens of the luminous component is also set in sections, focusing on each area to meet the imaging requirements of each section.

[0038] Each section of fill light can be controlled by gradual enhancement to meet the gradual adaptation process of the human eye to the environment. The wavelength of electromagnetic waves that the average human eye can perceive is between 400 and 760nm, but some people can perceive electromagnetic waves with a wavelength of about 380 to 780nm. Waves above 760nm are infrared waves, and those below 10nm to 400nm are ultraviolet waves. Some windshields and explosion-proof films will isolate the entry of infrared and ultraviolet waves. The overall transmittance of the front of the motor vehicle is required to be no less than 70% in the visible light band.

[0039] Ultra-long distance fill light wavelength λ ir Infrared band; long-distance fill light wavelength λ f The wavelength of the visible light component λ is within the green-yellow range of the visible light band, which is the part of the visible light band that the human eye is more sensitive to. Its luminous intensity is greater than the visible threshold of the human eye and less than the maximum value that the human eye can tolerate. m There are two control modes, the first wavelength λ m1 is the visible light band, yellow-red light range, λ m1 ≠λ f , the luminous intensity is greater than the long-distance fill light intensity and less than the maximum value that the human eye can tolerate; the second wavelength λ m2 It is a mixed wavelength of visible light and ultraviolet band, blue-violet light range (the longer the wavelength, the weaker the penetration, the shorter the wavelength, the stronger the penetration, and the shortest wavelength of purple light is more conducive to penetrating the window glass). Its luminous intensity can meet the imaging requirements of personnel and characteristic targets in the vehicle driver's space; short-range fill light wavelength λ n =λ m2 The luminous effect should be able to meet the imaging requirements of personnel and characteristic targets in the vehicle driver and passenger space; the illumination value of each target section gradually increases from far to near.

[0040] The video analysis component has the ability to perceive and analyze the illumination of imaging, and can output two classification results of sufficient illumination and insufficient illumination. It also has the function of identifying the vehicle model, license plate, vehicle logo and other external components based on infrared band imaging and visible light band imaging. The infrared imaging identifies the height of the driver and passenger position from the ground and outputs them in three categories: low (example: small car / sedan, about 1m), medium (example: medium-sized car / van, about 1.5m), and high (example: large car / truck / bus, about 3-3.5m);

[0041] The video analysis component can mark the capture time of the first appearance of the infrared fill light imaging wavelength characteristics, the long-distance visible light fill light imaging wavelength characteristics, the medium-distance visible light fill light imaging wavelength characteristics, and the close-distance visible light fill light imaging characteristics according to the timing information, and calculate the vehicle's driving time T and driving speed V (V=S / T), where S is the distance difference between the farthest position of the infrared fill light, the farthest position of the long-distance visible light, the farthest position of the medium-distance visible light, and the farthest position of the close-distance fill light.

[0042] The control method of the low-light fill-light capture system for a moving vehicle comprises the following steps:

[0043] S01: The video acquisition component captures the visible light image information of the vehicle target, and analyzes it through the video analysis component. When the output result is insufficient light, the main control component controls to turn on the ultra-long-distance infrared fill light;

[0044] S02: The video acquisition component captures the imaging information of the infrared characteristics of the vehicle target, records the imaging time t0, analyzes it through the video analysis component, classifies and outputs the vehicle type and the height of the driver and passenger position from the ground, and the main control component controls to turn on the long-distance fill light;

[0045] S03: The video acquisition component captures the visible light image information of the vehicle target under the long-distance fill light illumination feature, records the imaging time t1, calculates the vehicle speed V from time t0 to time t1 (t0-t1) through the video analysis component, and measures the vehicle speed V at L m The running time T in the range,

[0046] When the driver and passenger classification output is medium or high, within T time, the main control component controls to turn on and select the first type of medium-distance fill light, and the luminous brightness is gradually increased according to the vehicle speed ratio;

[0047] When the driver and passenger classification output is low, within T time, the main control component selects the first type of medium-distance fill light, and the luminous brightness is gradually increased according to the vehicle speed ratio, and then controls to turn on and select the second type of medium-distance fill light to achieve short-term lighting (flash).

[0048] Under the first fill light illumination condition, after color filtering compensation is performed by the image synthesis component, the video analysis component obtains structured information such as vehicle model and color;

[0049] Under the second fill light illumination condition, after color filtering compensation by the image synthesis component, the video analysis component obtains structural information such as faces and objects inside the vehicle driver and passenger position;

[0050] S04: The video acquisition component captures the visible light image information of the vehicle target under the mid-range fill light illumination characteristics, records the imaging time t2, and when the driver and passenger classification output is medium or high, calculates the vehicle speed V' and the focal length ratio from time t0 to time t2, and measures the time to enter L n Time, the main control component briefly turns on the short-distance fill light (flash) (similar to the flash of a flash, which can be single or continuous), and at the same time, after color filtering compensation by the image synthesis component, the video analysis component obtains structured information such as the face and objects inside the vehicle driver and passenger position;

[0051] S05: The video frame image with the highest definition is captured from each segmented video, and is superimposed and synthesized by an image synthesis component, and relevant structured information such as time, vehicle speed, vehicle color, model, and license plate is annotated.

[0052] The image component respectively converts the long-distance fill light illumination area image PIC-F, the medium-distance fill light illumination image PIC-M, and the short-distance fill light illumination image PIC-N into the image according to λ f , m , n The corresponding wavelengths are filtered for compensation (such as green-yellow, yellow-red and blue-violet visible light wavelengths), and then PIC-F, PIC-M and PIC-N are superimposed and synthesized according to the vehicle profile information obtained by the video analysis component; when the synthesis and superposition are inconsistent, the vehicle model and color shall be based on PIC-F, the license plate and logo graphics shall be based on PIC-M, and the driver and passenger appearance images shall be based on PIC-N.

[0053] The present invention adopts a multi-segment directional linear array controllable luminous lighting device, sets a multi-state mode of different fill light band lighting according to the distance of the moving vehicle to the capture pole position, and uses a video acquisition component to perform infrared, visible light and ultraviolet full-band perception imaging; the system classifies the vehicle model, driver and passenger height, etc. of the moving vehicle through a video analysis component and calculates the vehicle speed, and controls the fill light device through a main control component, and finally performs image filtering and compensation processing for the specific wavelength of the fill light to complete the multi-information synthesis and superposition of the captured picture. The system actively senses the position classification of the moving vehicle, and solves the problem of energy-saving sensing control of the fill light for the moving vehicle capture and the driver being affected by the fill light through a multi-band extended-range fill light mode; it improves the vehicle model adaptability problem of the fill light capture lighting of the moving vehicle.

[0054] In a low-illuminance environment, the present invention actively performs segmented extended-range fill-light illumination on the captured target through an arc-shaped multi-band directional line array fill-light device, and combines the infrared, visible light and ultraviolet full-band perception imaging and intelligent analysis and image color filtering compensation of the video acquisition component to solve the problems of energy-saving sensing control of the fill-light for capturing moving vehicles and the influence of the fill-light on the driver; and improves the vehicle model adaptability problem of fill-light capture illumination for moving vehicles.

[0055] The present invention is described in detail above in conjunction with the embodiments, but the contents are only specific implementations of the present invention, and should not be understood as limiting the scope of the patent of the present invention. It should be pointed out that for ordinary technicians in this field, without departing from the concept of the present invention, all modifications and improvements made according to the scope of the application of the present invention should still fall within the scope of the patent of the present invention.

Claims

1. A low-light fill-in capture system for a moving vehicle, characterized by: The system includes a fill light component, a video acquisition component, a video analysis component, an image synthesis component, a main control component and a timing position component; The output end of the video acquisition component is connected to the video analysis component and the image synthesis component respectively; the image synthesis component outputs to the main control component and the video analysis component respectively; the output end of the video analysis component and the time position component is connected to the main control component, and the main control component output controls the fill light component; The fill light component is an arc-shaped multi-band directional line array controllable fill light device, which is arranged in an arc shape toward the direction of vehicle travel to form a segmented capture fill light area. The vertical area of ​​the target illuminated by each segment of light is not less than 3 meters × 5 meters, with the fill light pole as the center and the height H; the distance between the pole and the illuminated target corresponds to the different focal length segments of the video acquisition component, which is L=f(x), where f(x) refers to the focal length and x refers to the range of each segmented focal length segment; the projection angle of each segment of the fill light illumination arc is θ=arc tan(L / H); the ultra-long-distance infrared light-emitting component is located at the upper part of the component, illuminating the ultra-long distance L ir Target: The long-distance visible light emitting component located in the upper middle part of the component illuminates the long-distance L f Target: The mid-range visible light emitting component located in the middle, illuminating the mid-range L m Target: The short-range visible light component located at the bottom of the component, irradiating the short distance L n Target: The focal length of the light-transmitting lens of the luminous component is also set in sections, focusing on each area to meet the imaging requirements of each section.

2. The low-light fill-in capture system for a moving vehicle according to claim 1, characterized in that: The video acquisition component adopts a fixed focal length, the infrared band focal length> the visible light back depth of field focal length> the visible light focal length> the visible light front depth of field focal length, and the video acquisition component has the imaging capability of sensing the infrared band, the entire visible light band and the ultraviolet band.

3. The low-light fill-in capture system for a moving vehicle according to claim 1, characterized in that: The ultra-long distance is 30 meters ≤ L ir ≤50 meters, long distance is 20 meters ≤L f ≤30 meters, medium distance is 10 meters ≤L m ≤20 meters, short distance is L n ≤10 meters.

4. The low-light fill-in capture system for a moving vehicle according to claim 3, characterized in that: Ultra-long distance fill light wavelength λ ir Infrared band; long-distance fill light wavelength λ f It is in the green-yellow light range in the visible light band; the wavelength of the medium-distance visible luminous component is λ m There are two control modes, the first wavelength λ m1 is the visible light band, yellow-red light range, λ m1 ≠λ f ; Type II wavelength λ m2 It is a mixed wavelength of visible light and ultraviolet band, blue-violet light range; close-range fill light wavelength λ n =λ m2 The luminous effect should be able to meet the imaging requirements of personnel and characteristic targets in the vehicle driver and passenger space; the illumination value of each target section gradually increases from far to near.

5. The low-light fill-in capture system for a moving vehicle according to claim 1, characterized in that: The video analysis component has the ability to perceive and analyze imaging illumination, and can output two classification results: sufficient lighting and insufficient lighting. It also has the function of identifying external components including vehicle model, license plate, and vehicle logo based on infrared band imaging and visible light band imaging. It can output the height of the driver and passengers above the ground in three categories based on infrared imaging: low, medium, and high.

6. The low-light fill-in capture system for a moving vehicle according to claim 4, characterized in that: The video analysis component can mark the capture time of the first appearance of the infrared fill light imaging wavelength characteristics, the long-distance visible light fill light imaging wavelength characteristics, the medium-distance visible light fill light imaging wavelength characteristics, and the close-distance visible light fill light imaging characteristics according to the timing information, and calculate the vehicle's driving time T and driving speed V, V=S / T, S is one of the distance differences among the farthest position of the infrared fill light, the farthest position of the long-distance visible light, the farthest position of the medium-distance visible light, and the farthest position of the close-distance fill light.

7. A control method for the low-light fill-in capture system for a moving vehicle according to any one of claims 1 to 6, characterized in that The steps include: S01: The video acquisition component captures the visible light image information of the vehicle target, and analyzes it through the video analysis component. When the output result is insufficient light, the main control component controls to turn on the ultra-long-distance infrared fill light; S02: The video acquisition component captures the imaging information of the infrared characteristics of the vehicle target, records the imaging time t0, analyzes it through the video analysis component, classifies and outputs the vehicle type and the height of the driver and passenger position from the ground, and the main control component controls to turn on the long-distance fill light; S03: The video acquisition component captures the visible light image information of the vehicle target under the long-distance fill light illumination characteristics, records the imaging time t1, calculates the vehicle speed V from time t0 to time t1 through the video analysis component, and measures the vehicle speed V at the medium distance L m The running time T in the range, When the driver and passenger classification output is medium or high, within T time, the main control component controls to turn on and select the first type of medium-distance fill light, and the luminous brightness is gradually increased according to the vehicle speed ratio; When the driver and passenger classification output is low, within T time, the main control component selects the first type of mid-range fill light, and the luminous brightness is gradually increased according to the vehicle speed ratio, and then controls to turn on and select the second type of mid-range fill light; Under the first fill light illumination condition, after color filtering compensation is performed by the image synthesis component, the video analysis component obtains structured information including vehicle model and vehicle color; Under the second fill light illumination condition, after color filtering compensation by the image synthesis component, the video analysis component obtains structural information including the position of the driver and passengers in the vehicle, and the faces and objects inside the vehicle; S04: The video acquisition component captures the visible light image information of the vehicle target under the mid-range fill light illumination characteristics, records the imaging time t2, and when the driver and passenger classification output is medium or high, calculates the vehicle speed V' and the focal length ratio from time t0 to time t2, and measures the vehicle speed V' and the focal length ratio when entering the close distance L n At that time, the main control component turns on the close-range fill light, and at the same time, after color filtering compensation by the image synthesis component, the video analysis component obtains structured information including the position of the vehicle driver and passengers, and the faces and objects inside. S05: The image synthesis component synthesizes the long-distance fill light illumination image PIC-F, the medium-distance fill light illumination image PIC-M, and the short-distance fill light illumination image PIC-N according to λ f , m , n The corresponding wavelength is filtered and compensated, and then PIC-F, PIC-M, and PIC-N are superimposed and synthesized according to the vehicle profile information obtained by the video analysis component; when the synthesis and superposition are inconsistent, the vehicle model and color shall be based on PIC-F, the license plate and logo graphics shall be based on PIC-M, and the driver and passenger appearance images shall be based on PIC-N.

Images