A vehicle driving night vision auxiliary perception system

Abstract

The present application relates to the technical field of vehicle driving, and discloses a vehicle driving night vision auxiliary perception system, an image acquisition module extracts a vehicle road region according to a night vision infrared image of a vehicle, and divides the vehicle road region into a plurality of segmented vehicle road regions; a region classification module analyzes the segmented vehicle road regions and determines corresponding region categories, the region categories including an interference road region and a perception road region; a coefficient calculation module calculates a region obstacle coefficient based on a perception pixel value corresponding to each perception road region; and an auxiliary perception module calculates a front road obstacle coefficient based on a vehicle position and each perception road region, adjusts the night vision infrared image, and performs night vision danger auxiliary perception on the vehicle based on the adjusted night vision infrared image, so that the interference region and the perception region can be dynamically distinguished, a precise night vision infrared image can be determined in combination with multidimensional parameters, and precise obstacle risk perception can be realized, thereby providing more reliable and safer night vision assistance for a driver.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle driving, in particular to a vehicle driving night vision auxiliary perception system. BACKGROUND

[0002] With the rapid development of the automobile industry, night driving safety has become an important research topic in the field of intelligent transportation. Due to insufficient light and low visibility at night, it is difficult for drivers to accurately identify obstacles or other potential dangers on the road, resulting in a significantly higher rate of traffic accidents at night than during the day.

[0003] Traditional night vision auxiliary perception relies on vehicle lighting systems such as headlights, but the lighting effect of vehicle lighting systems is limited in extreme darkness or adverse weather conditions, and cannot provide sufficient visual support for drivers, lacking fine segmentation and classification of road areas, resulting in ineffective differentiation between interference areas (such as vegetation and buildings on both sides of the road) and effective perception areas (such as obstacles in the center of the road), thereby reducing the accuracy and practicality of the system, resulting in insufficient sensitivity and robustness of obstacle detection, and difficulty in adapting to complex and variable night driving scenarios. SUMMARY

[0004] The present application provides a vehicle driving night vision auxiliary perception system, which can dynamically distinguish between interference areas and perception areas, determine accurate night vision infrared images based on multi-dimensional parameters, and further realize accurate obstacle risk perception, providing more reliable and safer night vision assistance for drivers.

[0005] To achieve the above purpose, the present application provides a vehicle driving night vision auxiliary perception system, comprising:

[0006] An image acquisition module is configured to acquire a night vision infrared image corresponding to the current time of the vehicle, extract a vehicle road area, and divide the vehicle road area into a plurality of segmented vehicle road areas;

[0007] A region classification module is configured to analyze each segmented vehicle road area and determine the region category corresponding to each segmented vehicle road area, wherein the region category includes an interference road region and a perception road region;

[0008] A coefficient calculation module is configured to extract each perception road region and calculate the region obstacle coefficient corresponding to each perception road region based on the corresponding perception pixel value within each perception road region;

[0009] An auxiliary perception module is configured to determine a vehicle position of the vehicle, calculate a front road obstacle coefficient of the vehicle according to the vehicle position and each segmented vehicle road region, and adjust the night vision infrared image according to the front road obstacle coefficient, and perform night vision danger auxiliary perception on the vehicle based on the adjusted night vision infrared image.

[0010] Further, the image acquisition module is configured to:

[0011] The image acquisition module is configured to acquire raw image data of a road scene based on a vehicle-mounted infrared camera corresponding to the vehicle;

[0012] The image acquisition module is configured to perform image preprocessing on the raw image data to obtain a corresponding night vision infrared image, wherein the image preprocessing includes non-uniformity correction and grayscale stretching.

[0013] Further, the region classification module is configured to:

[0014] The region classification module is configured to extract pixel values of the segmented vehicle road region to determine a corresponding region pixel value sequence;

[0015] The region classification module is configured to determine a previous time of the current time, and determine a previous pixel value sequence corresponding to the previous time;

[0016] The region classification module is configured to determine an image change factor of the segmented vehicle road region based on the region pixel value sequence and the previous pixel value sequence;

[0017] The region classification module is configured to pre-set a preset image change factor, and determine a region category corresponding to each segmented vehicle road region according to a relationship between the image change factor and the preset image change factor;

[0018] The region classification module is configured to classify the corresponding segmented vehicle road region as an interference road region when the image change factor is less than the preset image change factor;

[0019] The region classification module is configured to classify the corresponding segmented vehicle road region as a sensing road region when the image change factor is greater than or equal to the preset image change factor.

[0020] Further, the region classification module is configured to:

[0021] The region classification module is configured to one-to-one correspond match pixel values in the region pixel value sequence and the previous pixel value sequence to obtain a plurality of corresponding pixel value matching pairs;

[0022] The region classification module is configured to calculate a pixel value deviation of each pixel value matching pair, wherein the pixel value deviation is an absolute value of a difference between two pixel values in the pixel value matching pair;

[0023] The region classification module is configured to determine a maximum pixel value deviation from all the pixel value deviations;

[0024] The region classification module is configured to extract a minimum pixel value from the previous pixel value sequence and a maximum pixel value from the region pixel value sequence;

[0025] The region classification module is configured to determine an extreme pixel value deviation of the minimum pixel value and the maximum pixel value, wherein the extreme pixel value deviation is a difference between the maximum pixel value and the minimum pixel value;

[0026] The region classification module is configured to take a ratio of the extreme pixel value deviation to the pixel value deviation as an image change factor for segmenting a vehicle road region.

[0027] Further, the coefficient calculation module is configured to:

[0028] The coefficient calculation module is configured to construct a perception pixel value fitting curve according to corresponding perception pixel values in each perception road region, and determine a plurality of closed perception pixel value regions corresponding to the perception pixel value fitting curve;

[0029] The coefficient calculation module is configured to calculate a region obstacle coefficient corresponding to the perception road region based on all the closed perception pixel value regions.

[0030] Further, the coefficient calculation module is configured to:

[0031] The coefficient calculation module is configured to perform curve fitting on corresponding perception pixel values in each perception road region to obtain a perception pixel value fitting curve;

[0032] The coefficient calculation module is configured to determine an initial perception pixel value and a tail perception pixel value corresponding to the perception pixel value fitting curve;

[0033] The coefficient calculation module is configured to connect the initial perception pixel value and the tail perception pixel value to obtain a curve connection line;

[0034] The coefficient calculation module is configured to obtain a plurality of closed perception pixel value regions based on the curve connection line and the perception pixel value fitting curve.

[0035] Further, the coefficient calculation module is configured to:

[0036] The coefficient calculation module is configured to determine a corresponding standard perception pixel value in the perception road area, wherein the standard perception pixel value is a mean value corresponding to the perception road area.

[0037] The coefficient calculation module is configured to mark the standard perception pixel value on the perception pixel value fitting curve.

[0038] The coefficient calculation module is configured to determine a region barycenter corresponding to each closed perception pixel value region, and generate a region mark for each region barycenter according to the position of the region barycenter and the standard perception pixel value.

[0039] The coefficient calculation module is configured to generate a first region mark for the corresponding region barycenter when the region barycenter is less than the standard perception pixel value.

[0040] The coefficient calculation module is configured to generate a second region mark for the corresponding region barycenter when the region barycenter is equal to the standard perception pixel value.

[0041] The coefficient calculation module is configured to generate a third region mark for the corresponding region barycenter when the region barycenter is greater than the standard perception pixel value.

[0042] The coefficient calculation module is configured to calculate a region obstacle coefficient corresponding to the perception road area based on the first region mark, the second region mark, and the third region mark.

[0043] Further, the coefficient calculation module is configured to:

[0044] The coefficient calculation module is configured to count a first mark number of the first region mark, count a second mark number of the second region mark, and count a third mark number of the third region mark.

[0045] The region obstacle coefficient corresponding to the perception road area is calculated according to the following formula:

[0046] ;

[0047] wherein q is the region obstacle coefficient corresponding to the perception road area, w1 is the first mark number, w2 is the second mark number, w3 is the third mark number, and max is a maximum value symbol.

[0048] Further, the auxiliary perception module is configured to:

[0049] The auxiliary perception module is configured to determine a shortest road distance between each perception road area and the vehicle position.

[0050] The auxiliary perception module is configured to calculate a product value of each shortest road distance and a corresponding regional obstacle coefficient as a relative obstacle coefficient respectively;

[0051] The auxiliary perception module is configured to perform weighted average summation on the relative obstacle coefficients corresponding to each sensing road region to obtain a front road obstacle coefficient of the vehicle.

[0052] Further, the auxiliary perception module is configured to:

[0053] The auxiliary perception module is configured to obtain a current image enhancement adjustment value of the night vision infrared image;

[0054] The auxiliary perception module is configured to pre-set a first preset front road obstacle coefficient and a second preset front road obstacle coefficient;

[0055] The auxiliary perception module is configured to pre-set a first preset adjustment value, a second preset adjustment value and a third preset adjustment value;

[0056] The auxiliary perception module is configured to, when the front road obstacle coefficient is less than the first preset front road obstacle coefficient, calculate a first product value of the first preset adjustment value and the current image enhancement adjustment value, and adjust the night vision infrared image based on the first product value to obtain an adjusted night vision infrared image;

[0057] The auxiliary perception module is configured to, when the front road obstacle coefficient is greater than or equal to the first preset front road obstacle coefficient and less than the second preset front road obstacle coefficient, calculate a second product value of the second preset adjustment value and the current image enhancement adjustment value, and adjust the night vision infrared image based on the second product value to obtain an adjusted night vision infrared image;

[0058] The auxiliary perception module is configured to, when the front road obstacle coefficient is greater than or equal to the second preset front road obstacle coefficient, calculate a third product value of the third preset adjustment value and the current image enhancement adjustment value, and adjust the night vision infrared image based on the third product value to obtain an adjusted night vision infrared image.

[0059] Compared with the prior art, the present application has the following advantages:

[0060] The application discloses a vehicle driving night vision auxiliary perception system, an image acquisition module extracts a vehicle road region according to a night vision infrared image of a vehicle, and divides the vehicle road region into a plurality of segmented vehicle road regions; a region classification module analyzes the segmented vehicle road regions, and determines corresponding region categories, the region categories including an interference road region and a perception road region; a coefficient calculation module calculates a region obstacle coefficient based on a perception pixel value corresponding to each perception road region; and an auxiliary perception module calculates a front road obstacle coefficient based on a vehicle position and each perception road region, adjusts the night vision infrared image, and performs night vision danger auxiliary perception on the vehicle based on the adjusted night vision infrared image, so that the interference region and the perception region can be dynamically distinguished, the accurate night vision infrared image is determined in combination with multidimensional parameters, and then the accurate obstacle risk perception is realized, and more reliable and safer night vision assistance is provided for a driver. BRIEF DESCRIPTION OF DRAWINGS

[0061] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The detailed description is made with reference to the accompanying drawings.

[0062] Figure 1 A structure schematic diagram of a vehicle driving night vision auxiliary perception system in an embodiment of the application is shown. DETAILED DESCRIPTION

[0063] The specific embodiments of the application will be further described in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the application, but are not used to limit the scope of the application.

[0064] In the description of the present application, it should be understood that the terms "center", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the purpose of facilitating the description of the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application.

[0065] The terms "first", "second", "third", etc. are only used for description purpose, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the technical features referred to. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, unless otherwise specified, the meaning of "a plurality of" is two or more.

[0066] In the description of the present application, it should be noted that unless otherwise explicitly specified and limited, the terms "mounting", "connection", "connecting" should be understood in a broad sense, for example, can be fixed connection, can also be detachable connection, or integral connection; can be mechanical connection, can also be electrical connection; can be direct connection, can also be indirect connection through an intermediate medium, can be internal communication of two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0067] The following is a description of the preferred embodiments of the present application in conjunction with the accompanying drawings.

[0068] As shown in Figure 1 Embodiments of the present application disclose a vehicle driving night vision auxiliary perception system, comprising:

[0069] An image acquisition module is configured to acquire a night vision infrared image corresponding to the vehicle at the current time, extract a vehicle road region, and divide the vehicle road region into a plurality of segmented vehicle road regions.

[0070] In the present embodiment, the extracted vehicle road region is cut into a plurality of small blocks as segmented vehicle road regions. For example, the division can be performed according to a fixed size grid, which is not specifically limited here.

[0071] In some embodiments of the present application, the image acquisition module is configured to:

[0072] The image acquisition module is configured to acquire raw image data of a road scene based on a vehicle-mounted infrared camera corresponding to the vehicle.

[0073] The image acquisition module is configured to perform image preprocessing on the raw image data to obtain a corresponding night vision infrared image, wherein the image preprocessing includes non-uniformity correction and gray scale stretching.

[0074] In the present embodiment, the raw image data refers to an image directly acquired by the vehicle-mounted infrared camera, and the non-uniformity correction and gray scale stretching are means for clear processing of the raw image data, which will not be introduced again here.

[0075] The beneficial effects of the above technical solution are: the present application performs non-uniformity correction and gray scale stretching on the raw image data, which can further ensure the clarity of the image.

[0076] A region classification module is configured to analyze each segmented vehicle road region and determine a region category corresponding to each segmented vehicle road region, wherein the region category includes an interference road region and a sensing road region.

[0077] In some embodiments of the present application, the region classification module is configured to:

[0078] The region classification module is configured to extract pixel values of the segmented vehicle road region, and determine a corresponding region pixel value sequence;

[0079] The region classification module is configured to determine a previous time corresponding to the current time, and determine a previous pixel value sequence corresponding to the previous time;

[0080] The region classification module is configured to determine an image change factor of the segmented vehicle road region based on the region pixel value sequence and the previous pixel value sequence;

[0081] The region classification module is configured to pre-set a preset image change factor, and determine a region category corresponding to each segmented vehicle road region according to a relationship between the image change factor and the preset image change factor;

[0082] The region classification module is configured to classify the corresponding segmented vehicle road region as an interference road region when the image change factor is less than the preset image change factor;

[0083] The region classification module is configured to classify the corresponding segmented vehicle road region as a sensing road region when the image change factor is greater than or equal to the preset image change factor.

[0084] In this embodiment, the previous time is a historical time, and the previous time is a time point including both pedestrians and vehicles, thereby ensuring that the determination of the sensing road region is based on a real dynamic background.

[0085] In this embodiment, when determining the previous pixel value sequence corresponding to the previous time, a previous vehicle road region corresponding to the previous time is determined, a plurality of previous segmented vehicle road regions are determined, and pixel values are determined. The specific manner is consistent with the determination manner of the region pixel value sequence of the current time, which is not repeated here.

[0086] In this embodiment, the image change factor can represent the change of the segmented vehicle road region.

[0087] In this embodiment, the preset image change factor is preferably 5, and can also be adjusted according to actual conditions.

[0088] In this embodiment, the interference road region includes flowers, grass, trees, buildings and the like.

[0089] In this embodiment, the sensing road region includes pedestrians and vehicles.

[0090] In this embodiment, based on the last pixel value sequence corresponding to the last moment, the interference road region and the sensing road region corresponding to the current moment can be determined, that is, the discrimination of the interference road region and the sensing road region is based on the real-time change of the current moment. The application provides a determination method for one moment, and the specific adaptive determination for future moments can be made.

[0091] The beneficial effects of the above technical solutions are: the application determines the interference road region and the sensing road region corresponding to each segmented vehicle road region according to the relationship between the image change factor and the preset image change factor, ensures the intelligence and rapidity of judgment, avoids the existing human eye recognition method, lays a foundation for night vision auxiliary perception, improves the accuracy and practicability of night vision auxiliary perception, and avoids the perception of false obstacles.

[0092] In some embodiments of the application, the region classification module is configured to:

[0093] The region classification module is configured to one-to-one corresponding matching of the pixel values in the region pixel value sequence and the last pixel value sequence, to obtain a plurality of corresponding pixel value matching pairs;

[0094] The region classification module is configured to calculate the pixel value deviation corresponding to each pixel value matching pair, wherein the pixel value deviation is the absolute value of the difference between the two pixel values in the pixel value matching pair;

[0095] The region classification module is configured to determine the maximum pixel value deviation from all pixel value deviations;

[0096] The region classification module is configured to extract the minimum pixel value from the last pixel value sequence and the maximum pixel value from the region pixel value sequence;

[0097] The region classification module is configured to determine the extreme pixel value deviation of the minimum pixel value and the maximum pixel value, wherein the extreme pixel value deviation is the difference between the maximum pixel value and the minimum pixel value;

[0098] The region classification module is configured to take the ratio of the extreme pixel value deviation to the pixel value deviation as the image change factor of the segmented vehicle road region.

[0099] In this embodiment, the pixel values in the region pixel value sequence and the last pixel value sequence are one-to-one corresponding matched, for example, the pixel value corresponding to a certain position of the segmented vehicle road region is 120, and the pixel value corresponding to the same position is extracted from the last pixel value sequence, for example, 115, and then the two pixel values are matched.

[0100] The beneficial effects of the above technical solutions are: the region pixel value sequence and the pixel values in the last pixel value sequence are one-to-one corresponding matched, the accuracy of road region change analysis can be ensured, errors can be avoided, the extreme pixel value deviation of the minimum pixel value and the maximum pixel value is determined, the maximum change of the road region can be determined, the ratio of the extreme pixel value deviation to the pixel value deviation is taken as an image change factor of the segmented vehicle road region, and then the image change factor of the segmented vehicle road region is obtained, the accuracy and comprehensiveness of the road change analysis are ensured, and reliable data support is provided for the front road obstacle analysis.

[0101] The coefficient calculation module is configured to extract each sensing road region, and calculate a region obstacle coefficient corresponding to each sensing road region based on a corresponding sensing pixel value in each sensing road region.

[0102] In the embodiment, for convenience of distinction, the corresponding pixel value in the sensing road region is taken as a sensing pixel value.

[0103] In some embodiments of the present application, the coefficient calculation module is configured to:

[0104] The coefficient calculation module is configured to construct a sensing pixel value fitting curve according to the corresponding sensing pixel value in each sensing road region, and determine a plurality of closed sensing pixel value regions corresponding to the sensing pixel value fitting curve.

[0105] The coefficient calculation module is configured to calculate the region obstacle coefficient corresponding to the sensing road region based on all the closed sensing pixel value regions.

[0106] In some embodiments of the present application, the coefficient calculation module is configured to:

[0107] The coefficient calculation module is configured to perform curve fitting on the corresponding sensing pixel value in each sensing road region to obtain a sensing pixel value fitting curve.

[0108] The coefficient calculation module is configured to determine an initial sensing pixel value and a tail sensing pixel value corresponding to the sensing pixel value fitting curve.

[0109] The coefficient calculation module is configured to connect the initial sensing pixel value and the tail sensing pixel value to obtain a curve connection line.

[0110] The coefficient calculation module is configured to obtain a plurality of closed sensing pixel value regions based on the curve connection line and the sensing pixel value fitting curve.

[0111] In the embodiment, the initial perception pixel value and the end perception pixel value are determined, and a curve connection line is obtained by connecting the two points, the curve connection line passes through the perception pixel value fitting curve, and a plurality of closed perception pixel value regions are obtained.

[0112] The beneficial effects of the above technical solutions are: the plurality of closed perception pixel value regions are obtained based on the curve connection line and the perception pixel value fitting curve, the fine division of the perception pixel value is realized, the perception pixel value in the change range is integrated and analyzed, and the accuracy of the night vision auxiliary perception of the vehicle driving is effectively ensured.

[0113] In some embodiments of the present application, the coefficient calculation module is configured to:

[0114] The coefficient calculation module is configured to determine the corresponding standard perception pixel value in the perception road area, wherein the standard perception pixel value is the mean value corresponding to the perception road area;

[0115] The coefficient calculation module is configured to mark the standard perception pixel value on the perception pixel value fitting curve;

[0116] The coefficient calculation module is configured to determine the region gravity center corresponding to each closed perception pixel value region, and generate a region mark for each region gravity center according to the position of each region gravity center and the standard perception pixel value;

[0117] The coefficient calculation module is configured to generate a first region mark for the corresponding region gravity center when the region gravity center is less than the standard perception pixel value;

[0118] The coefficient calculation module is configured to generate a second region mark for the corresponding region gravity center when the region gravity center is equal to the standard perception pixel value;

[0119] The coefficient calculation module is configured to generate a third region mark for the corresponding region gravity center when the region gravity center is greater than the standard perception pixel value;

[0120] The coefficient calculation module is configured to calculate the region obstacle coefficient corresponding to the perception road area based on the first region mark, the second region mark and the third region mark.

[0121] In the embodiment, the standard perception pixel value is marked on the perception pixel value fitting curve, that is, the position of the standard perception pixel value is determined.

[0122] In the embodiment, different marks are generated according to the positions of the region gravity centers.

[0123] The beneficial effects of the above technical solutions are: the application calculates the region obstacle coefficient corresponding to the perceived road region based on the first region mark, the second region mark and the third region mark, when the region barycenter is less than, equal to or greater than the standard perceived pixel value, the obstacle conditions represented are different, therefore, region marking is performed here, and then the region obstacle change state is fed back.

[0124] In some embodiments of the present application, the coefficient calculation module is configured to:

[0125] The coefficient calculation module is configured to count the first mark quantity of the first region mark, count the second mark quantity of the second region mark, and count the third mark quantity of the third region mark.

[0126] The region obstacle coefficient corresponding to the perceived road region is calculated according to the following formula:

[0127]

[0128] Wherein, q is the region obstacle coefficient corresponding to the perceived road region, w1 is the first mark quantity, w2 is the second mark quantity, w3 is the third mark quantity, and max is the maximum value symbol.

[0129] In this embodiment, max (w2, w3) means selecting the maximum value from w2 and w3, and if they are the same, one can be randomly selected.

[0130] The auxiliary perception module is configured to determine the vehicle position of the vehicle, calculate the front road obstacle coefficient of the vehicle according to the vehicle position and each perceived road region, adjust the night vision infrared image according to the front road obstacle coefficient, and perform night vision danger auxiliary perception on the vehicle based on the adjusted night vision infrared image.

[0131] In some embodiments of the present application, the auxiliary perception module is configured to:

[0132] The auxiliary perception module is configured to determine the shortest road distance between each perceived road region and the vehicle position.

[0133] The auxiliary perception module is configured to calculate the product value of each shortest road distance and the corresponding region obstacle coefficient respectively as a relative obstacle coefficient.

[0134] The auxiliary perception module is configured to perform weighted average summation on the relative obstacle coefficient corresponding to each perceived road region to obtain the front road obstacle coefficient of the vehicle.

[0135] In this embodiment, when determining the shortest road distance, the spatial coordinates of the perceived road region and the spatial coordinates of the vehicle position can be used for determination. ​

[0136] In the embodiment, different weights are given to each sensing road area, and the weights can be given according to a subjective weighting method or an objective weighting method.

[0137] The above technical solution has the following beneficial effects: the relative obstacle coefficients corresponding to each sensing road area are weighted and averaged to obtain the front road obstacle coefficient of the vehicle, and the relative existence of the vehicle and the front obstacle can be fed back through the front road obstacle coefficient, thereby laying a better foundation for image enhancement processing.

[0138] In some embodiments of the present application, the auxiliary perception module is configured to:

[0139] The auxiliary perception module is configured to obtain a current image enhancement adjustment value of the night vision infrared image.

[0140] The auxiliary perception module is configured to pre-set a first preset front road obstacle coefficient and a second preset front road obstacle coefficient.

[0141] The auxiliary perception module is configured to pre-set a first preset adjustment value, a second preset adjustment value, and a third preset adjustment value.

[0142] When the front road obstacle coefficient is less than the first preset front road obstacle coefficient, the auxiliary perception module is configured to calculate a first product value of the first preset adjustment value and the current image enhancement adjustment value, and adjust the night vision infrared image based on the first product value to obtain an adjusted night vision infrared image.

[0143] When the front road obstacle coefficient is greater than or equal to the first preset front road obstacle coefficient and less than the second preset front road obstacle coefficient, the auxiliary perception module is configured to calculate a second product value of the second preset adjustment value and the current image enhancement adjustment value, and adjust the night vision infrared image based on the second product value to obtain an adjusted night vision infrared image.

[0144] When the front road obstacle coefficient is greater than or equal to the second preset front road obstacle coefficient, the auxiliary perception module is configured to calculate a third product value of the third preset adjustment value and the current image enhancement adjustment value, and adjust the night vision infrared image based on the third product value to obtain an adjusted night vision infrared image.

[0145] In the embodiment, the current image enhancement adjustment value is set according to the dynamic range and signal-to-noise ratio performance of the vehicle-mounted infrared camera. The current image enhancement adjustment value can enhance the contrast, brightness, and sharpness of the image.

[0146] In the embodiment, the first preset front road obstacle coefficient is preferably 12, and the second preset front road obstacle coefficient is preferably 16, and the actual demand can be adaptively adjusted.

[0147] In the embodiment, the first preset adjustment value is preferably 1.15, the second preset adjustment value is preferably 1.25, and the third preset adjustment value is preferably 1.35.

[0148] In the embodiment, the night vision infrared image is enhanced and processed according to the first product value, the second product value or the third product value, and an adjusted night vision infrared image is obtained.

[0149] The beneficial effects of the above technical solution are: according to the front road obstacle coefficient, the first preset front road obstacle coefficient and the second preset front road obstacle coefficient, the corresponding preset adjustment value is selected, the adjustment of the current image enhancement adjustment value is realized, and the enhancement processing of the night vision infrared image is further realized, the enhancement information can be provided for the image in the night vision scene, the visual perception ability in the night vision is improved, the noise point interference can be reduced, the obstacle contour can be enhanced, the adaptability and reliability are higher, and the risk of night driving accidents can be effectively reduced.

[0150] In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0151] Although the present application has been described with reference to the embodiments above, various modifications can be made to it without departing from the scope of the present application and equivalent components can be substituted therefor. In particular, each feature in the embodiments disclosed in the present specification can be combined with any other features in any manner, provided that there is no structural conflict. The combinations of these features are not all described in the present specification only for the purpose of saving space and resources.

[0152] Those skilled in the art can understand that the above are only preferred embodiments of the present application, and are not used to limit the present application, although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent substitutions for part of the technical features. Any modification, equivalent substitution, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A vehicle drive night vision assist perception system, comprising: The method comprises the following steps: An image acquisition module is configured to acquire a night-vision infrared image corresponding to a current time of a vehicle, extract a vehicle road region, and divide the vehicle road region into a plurality of segmented vehicle road regions; A region classification module is configured to analyze each segmented vehicle road region and determine a region category corresponding to each segmented vehicle road region, wherein the region category comprises an interference road region and a perception road region; A coefficient calculation module is configured to extract each perception road region, and calculate a region obstacle coefficient corresponding to each perception road region based on a corresponding perception pixel value in each perception road region; The coefficient calculation module is configured to: fit the corresponding perception pixel value in each perception road region to obtain a perception pixel value fitting curve; determine an initial perception pixel value and a tail perception pixel value corresponding to the perception pixel value fitting curve; connect the initial perception pixel value and the tail perception pixel value to obtain a curve connection line; obtain a plurality of closed perception pixel value regions based on the curve connection line and the perception pixel value fitting curve; determine a standard perception pixel value corresponding to the perception road region, wherein the standard perception pixel value is a mean value corresponding to the perception road region; mark the standard perception pixel value on the perception pixel value fitting curve; determine a region barycenter corresponding to each closed perception pixel value region, and generate a region mark for each region barycenter according to the position of each region barycenter and the standard perception pixel value; generate a first region mark for the corresponding region barycenter when the region barycenter is less than the standard perception pixel value; generate a second region mark for the corresponding region barycenter when the region barycenter is equal to the standard perception pixel value; generate a third region mark for the corresponding region barycenter when the region barycenter is greater than the standard perception pixel value; calculate the region obstacle coefficient corresponding to the perception road region based on the first region mark, the second region mark, and the third region mark; An auxiliary perception module is configured to determine a vehicle position of the vehicle, calculate a front road obstacle coefficient of the vehicle based on the vehicle position and each perception road region, adjust the night-vision infrared image based on the front road obstacle coefficient, and perform night-vision danger auxiliary perception on the vehicle based on the adjusted night-vision infrared image.

2. The vehicle drive night vision assist perception system of claim 1, wherein, The image acquisition module is configured to: The image acquisition module is configured to acquire original image data of a road scene based on a vehicle-mounted infrared camera corresponding to the vehicle; The image acquisition module is configured to perform image preprocessing on the original image data to obtain a corresponding night-vision infrared image, wherein the image preprocessing comprises non-uniformity correction and gray scale stretching.

3. The vehicle drive night vision assist perception system of claim 1, wherein, The region classification module is configured to: The region classification module is configured to extract a region pixel value sequence corresponding to the segmented vehicle road region; The region classification module is configured to determine a previous time of the current time, and determine a previous pixel value sequence corresponding to the previous time. The region classification module is configured to determine an image change factor of the segmented vehicle road region based on the region pixel value sequence and the previous pixel value sequence; The region classification module is configured to pre-set a preset image change factor, and determine a region category corresponding to each segmented vehicle road region according to a relationship between the image change factor and the preset image change factor; The region classification module is configured to classify the corresponding segmented vehicle road region as an interference road region when the image change factor is less than the preset image change factor; The region classification module is configured to classify the corresponding segmented vehicle road region as a sensitive road region when the image change factor is greater than or equal to the preset image change factor.

4. The vehicle drive night vision assist perception system of claim 3, wherein, The region classification module is configured to: The region classification module is configured to perform one-to-one corresponding matching on the pixel values in the region pixel value sequence and the previous pixel value sequence to obtain a plurality of corresponding pixel value matching pairs; The region classification module is configured to calculate a pixel value deviation corresponding to each pixel value matching pair, wherein the pixel value deviation is an absolute value of a difference between the two pixel values in the pixel value matching pair; The region classification module is configured to determine a maximum pixel value deviation from all the pixel value deviations; The region classification module is configured to extract a minimum pixel value from the previous pixel value sequence and extract a maximum pixel value from the region pixel value sequence; The region classification module is configured to determine an extreme pixel value deviation of the minimum pixel value and the maximum pixel value, wherein the extreme pixel value deviation is a difference between the maximum pixel value and the minimum pixel value; The region classification module is configured to take a ratio of the extreme pixel value deviation to the pixel value deviation as an image change factor of the segmented vehicle road region.

5. The vehicle drive night vision assist perception system of claim 1, wherein, The coefficient calculation module is configured to: The coefficient calculation module is configured to count a first number of first region marks, count a second number of second region marks, and count a third number of third region marks; The region obstacle coefficient corresponding to the sensitive road region is calculated according to the following formula: ； wherein q is the region obstacle coefficient corresponding to the sensitive road region, w1 is the first number of marks, w2 is the second number of marks, w3 is the third number of marks, and max is a maximum value symbol.

6. The vehicle drive night vision assist perception system of claim 1, wherein, The auxiliary perception module is configured to: The auxiliary perception module is configured to determine a shortest road distance between each sensitive road region and the vehicle position; The auxiliary perception module is configured to calculate a product value of each shortest road distance and a corresponding region obstacle coefficient as a relative obstacle coefficient, respectively; The auxiliary perception module is configured to perform weighted average summation on the relative obstacle coefficients corresponding to each sensitive road region to obtain a front road obstacle coefficient of the vehicle.

7. The vehicle drive night vision assist perception system of claim 1, wherein, The auxiliary perception module is configured to: The auxiliary perception module is configured to obtain a current image enhancement adjustment value of the night vision infrared image; The auxiliary perception module is configured to pre-set a first preset front road obstacle coefficient and a second preset front road obstacle coefficient; The auxiliary perception module is configured to pre-set a first preset adjustment value, a second preset adjustment value, and a third preset adjustment value; The auxiliary perception module is configured to: The auxiliary perception module is configured to, when the front road obstacle coefficient is less than the first preset front road obstacle coefficient, calculate a first product value of the first preset adjustment value and the current image enhancement adjustment value, and adjust the night vision infrared image based on the first product value to obtain an adjusted night vision infrared image. The auxiliary perception module is configured to, when the front road obstacle coefficient is greater than or equal to the first preset front road obstacle coefficient and less than the second preset front road obstacle coefficient, calculate a second product value of the second preset adjustment value and the current image enhancement adjustment value, and adjust the night vision infrared image based on the second product value to obtain an adjusted night vision infrared image. The auxiliary perception module is configured to, when the front road obstacle coefficient is greater than or equal to the second preset front road obstacle coefficient, calculate a third product value of the third preset adjustment value and the current image enhancement adjustment value, and adjust the night vision infrared image based on the third product value to obtain an adjusted night vision infrared image.

Images