Intelligent highway maintenance marking quality detection method, system and electronic equipment

Abstract

The present application relates to an intelligent road maintenance marking quality detection method, system and electronic equipment, which includes obtaining image information at a preset vertical coordinate position in a frame stream of road markings according to a starting point trigger signal, and recording the image information in a cache library; triggering the image information in the combined cache library to be imaged according to a preset imaging rule, obtaining a two-dimensional plane map, and generating a road section unique code identifier for the two-dimensional plane image, through which the address information recorded in the frame stream image can be obtained; extracting and calculating the two-dimensional plane map to obtain a marking rectangular image; judging whether the marking rectangular image is qualified according to a preset judgment rule, and if unqualified, retaining the road section unique code identifier in the database. The present application has the effect of automatically inspecting and identifying the quality of road markings, thereby improving work efficiency.

Description

Technical Field

[0001] The present application relates to the field of image recognition technology, and in particular to an intelligent highway maintenance marking quality detection method, system and electronic equipment. Background Art

[0002] Traffic markings refer to the markings on the road surface that convey traffic information such as guidance, restriction, and warning to traffic participants. Their function is to control and guide traffic. Clear and conspicuous markings can better ensure traffic safety and guide civilized driving.

[0003] At present, in the process of inspecting highway markings, the status of the markings is generally judged by visual inspection. The current standard for when to re-mark the markings is non-quantitative, and different inspectors have different understandings of "good visibility", which sometimes results in repeated markings, resulting in a waste of financial resources, or no re-marking, which easily leads to potential traffic accidents. Therefore, the inventor believes that the current inspection marking standards are not unified, the level of automation is not high, the efficiency is low, and the level of intelligence is low, which needs to be further improved. Summary of the invention

[0004] In order to realize automated inspection and identification of road marking quality and thus improve work efficiency, the present application provides an intelligent road maintenance road marking quality detection method, system and electronic equipment.

[0005] In the first aspect, the present application provides an intelligent highway maintenance marking quality detection method, which adopts the following technical solution:

[0006] An intelligent highway maintenance marking quality detection method, comprising:

[0007] Acquire image information at a preset ordinate position in the highway marking frame stream according to a starting point trigger signal, and record the image information in a cache library;

[0008] According to the preset imaging rules, the image information in the combined cache library is triggered to be imaged to obtain a two-dimensional plane map, and a road section unique code identifier is generated for the two-dimensional plane image. The address information recorded in the frame stream image can be obtained through the road section unique code identifier;

[0009] Extract and calculate the two-dimensional plane image to obtain a marked line rectangular image;

[0010] The road marking rectangular image is judged to be qualified according to the preset judgment rules. If it is unqualified, the road section unique code is retained in the database.

[0011] By adopting the above technical scheme, the image information at the preset vertical coordinate position of the highway marking frame stream is obtained and recorded in the cache library, and the image information in the cache library is combined and imaged according to the preset imaging rules to obtain a two-dimensional plane map. The two-dimensional plane map has a road section unique code mark, and the address information of the recorded marking can be identified from the road section unique code mark. Then the two-dimensional plane map is calculated and processed to increase the accuracy of detection, and a rectangular image of the marking is obtained. It is judged whether the rectangular image of the marking is qualified according to the preset judgment rules, thereby realizing automatic and intelligent collection and identification of the quality of the marking, improving the efficiency of the operation, and compared with the manual inspection method through subjective observation, it avoids the inconsistent understanding of visibility by the inspectors, and improves the objectivity and accuracy of the inspection.

[0012] Optionally, the preset imaging rule includes:

[0013] After receiving the starting point trigger signal, the vehicle speed and time are recorded in real time, and the current distance value is calculated. When the current distance value reaches the preset distance threshold, the acquisition of the image information at the preset vertical coordinate position in the highway marking frame stream screen is stopped, and at the same time, the acquisition of the image information at the preset vertical coordinate position in the highway marking frame stream screen is restarted until the vehicle stop signal is received.

[0014] By adopting the above technical solution, segmented road marking collection and recognition can release memory in a timely manner compared to dynamic path recognition. As long as a section of road marking is qualified, the memory can be released to ensure that the cache space is continuously available, effectively enhancing the sustainability of patrol operations.

[0015] Optionally, the step of imaging the image information in the combined cache library to obtain a two-dimensional plane map includes:

[0016] Match the pixel values ​​of two adjacent columns recorded in the cache to obtain the similarity;

[0017] When the similarity is greater than the preset threshold, similar content is deleted according to the preset deletion rules;

[0018] The remaining pixel values ​​in the cache are combined in chronological order to obtain a two-dimensional plane image.

[0019] By adopting the above technical solution, the set similarity matching and preset deletion rules can effectively compress the image, reduce the stretching and deformation of the picture caused by the excessive frequency of many frame streams, and make the image more consistent with the normal visual two-dimensional plane map of the marking line, which is conducive to the identification of the marking quality.

[0020] Optionally, in the step of matching the pixel values ​​of two adjacent columns recorded in the cache to obtain the similarity, cosine similarity is used to calculate the similarity of two adjacent rows of pixel values ​​in the cache. The cosine similarity algorithm is as follows:

[0021]

[0022] Among them, A and B represent the pixel values ​​of two adjacent rows, A·B represents the inner product of the two rows of pixel values, and ||A||*||B|| represents the multiplication of the norms of the two rows of pixel values.

[0023] By adopting the above technical solution, the cosine similarity describes the similarity of changes in pixel values ​​of two rows. The result is more accurate than the general pixel difference algorithm, which is conducive to the realization of two-dimensional plane image imaging.

[0024] Optionally, the step of extracting and calculating the two-dimensional plane image to obtain a marking line image includes:

[0025] Preprocess the two-dimensional plane image to obtain an enhanced grayscale image;

[0026] Perform edge detection on the enhanced grayscale image to obtain the edge information of the marking line;

[0027] Performing Hough transform on the edge information of the marking line to identify the shape of the marking line, and extracting the shape of the marking line to obtain a marking line image;

[0028] Paste the marking line image into a layer of preset fixed size, and automatically fill the preset pixel value into the blank space of the layer to obtain the marking line rectangular image.

[0029] By adopting the above technical solution, preprocessing can effectively eliminate the influence of blur or different lighting of the two-dimensional plane image, convert it into an enhanced grayscale image, and then obtain the edge information of the marking line by performing edge detection on the enhanced grayscale image. Then, the shape of the marking line is identified by Hough transform to obtain the marking line image, and the marking line image is pasted into a layer with uniform pixel values. The preset pixel value is set with a contrast with the pixel value at the marking line, which reduces the influence of road surfaces other than the marking line on the recognition quality, thereby improving the accuracy of quality recognition.

[0030] Optionally, the step of preprocessing the two-dimensional plane image to obtain an enhanced grayscale image includes:

[0031] Perform an opening operation of first corroding and then dilating the two-dimensional plane image to obtain a primary processed image;

[0032] The primary processed image is grayed out by weighted average method to obtain a grayscale image;

[0033] Perform median filtering on the grayscale image to obtain an enhanced grayscale image.

[0034] By adopting the above technical solution, the opening operation has the effect of eliminating small objects, separating objects at thin places and smoothing the boundaries of larger objects. The median filter has a significant effect on salt and pepper noise and speckle noise, and the median filter has a good edge preservation characteristic, which can further enhance the distinction of the situation where the edges of the marking lines are incomplete.

[0035] Optionally, the preset judgment rule includes:

[0036] The number of pixel values ​​in the standard rectangular image that are less than a preset threshold is counted, and based on the ratio of the number to the total number, it is determined whether the ratio is greater than a preset ratio. If the ratio is greater than the preset ratio, the marked rectangular image is determined to be unqualified.

[0037] Since the marking pattern in the standard rectangular image is generally rectangular and not overly complex, the overall calculation processing speed can be effectively improved by counting the number ratios, and the cache can be released in a timely manner, thereby enhancing the duration of the quality recognition operation.

[0038] In the second aspect, the present application provides an intelligent highway maintenance marking quality detection system, which adopts the following technical solutions:

[0039] An intelligent highway maintenance marking quality detection system, comprising:

[0040] An information recording module, used to obtain image information at a preset ordinate position in a highway marking frame stream according to a starting point trigger signal, and record the image information in a cache library;

[0041] The information imaging module is used to trigger the image information in the combined cache library to be imaged according to the preset imaging rules, obtain a two-dimensional plane map, and generate a road section unique code identification for the two-dimensional plane image, and the address information recorded in the frame stream screen can be obtained through the road section unique code identification;

[0042] An image processing module is used to extract and calculate the two-dimensional plane image to obtain a marked rectangular image;

[0043] The image judgment module is used to judge whether the road marking rectangular image is qualified according to the preset judgment rules. If it is unqualified, the road section unique code identification is retained in the database.

[0044] In a third aspect, the present application provides an electronic device, which adopts the following technical solution:

[0045] An electronic device comprises a memory and a processor, wherein the memory stores a computer program which can be loaded by the processor and execute the above-mentioned intelligent highway maintenance line marking quality detection method.

[0046] In a fourth aspect, the present application provides a computer storage medium, which adopts the following technical solution:

[0047] A computer-readable storage medium stores a computer program that can be loaded by a processor and execute the above-mentioned intelligent highway maintenance marking quality detection method.

[0048] In summary, this application includes the following beneficial technical effects:

[0049] The image information at the preset vertical coordinate position of the highway marking frame stream is acquired and recorded in the cache library. The image information in the cache library is combined and imaged according to the preset imaging rules to obtain a two-dimensional plane map. The two-dimensional plane map has a road section unique code mark. The address information of the recorded marking can be identified from the road section unique code mark. The two-dimensional plane map is then calculated and processed to increase the accuracy of the detection, and a rectangular image of the marking is obtained. The qualified rectangular image of the marking is judged according to the preset judgment rules, thereby realizing automatic collection and identification of the quality of the marking, unifying the qualified identification standards, and improving the efficiency of the operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a flow chart of an intelligent highway maintenance marking quality detection method according to one embodiment of the present application.

[0051] Figure 2 It is a flowchart of the specific steps of combining the image information in the cache library to perform imaging in step S2 in another embodiment of the present application.

[0052] Figure 3 This is a specific step flow chart of step S3 in another embodiment of the present application.

[0053] Figure 4 It is a specific step flow chart of step S30 in another embodiment of the present application.

[0054] Figure 5 It is a structural diagram of an intelligent highway maintenance marking quality detection system according to one embodiment of the present application.

[0055] In the figure, 1. Information recording module; 2. Information imaging module; 3. Image processing module; 4. Image judgment module. DETAILED DESCRIPTION

[0056] The following is combined with Figure 1-5 This application is described in further detail.

[0057] In order to make the purpose, technical solutions and advantages of this application more clear, the following Figure 1-5It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0058] Reference Figure 1 The present application discloses an intelligent highway maintenance marking quality detection method, comprising the following steps:

[0059] S1. Obtain image information at a preset vertical coordinate position in a highway marking frame stream according to a starting point trigger signal, and record the image information in a cache library.

[0060] Specifically, by installing a data acquisition camera on the vehicle, the camera is fixedly set at the front of the vehicle and perpendicular to the road surface, and the ground can be clearly focused and photographed. The camera can be selected as a fisheye camera or an ordinary lens, and this method is not limited by the camera type. When it is necessary to inspect the highway markings, connect the vehicle navigation system, record the starting point address, and generate a starting point trigger signal through a preset button or voice command and send it to the server. After the server obtains the starting point trigger signal, it starts to obtain the image information at the preset vertical coordinate position in the highway marking frame stream through the camera. In this embodiment, the pixel value in the horizontal direction of the most middle area of ​​the vertical coordinate position of the frame stream is obtained first. This is because the most middle position of the frame stream captured by the camera is perpendicular to the highway marking, and the image quality is good. This position can effectively avoid the reflection of the road surface in the distance when the sun is strong. The impact on the frame stream image.

[0061] In addition, the cache can be of fixed length to ensure that the cache does not overflow during long-term recording. Specifically, the cache is responsible for recording the input vertical coordinate scanning results in the order of shooting. When the recorded content exceeds the maximum capacity of the cache, the first-in-first-out method is used to delete the early data from the cache and record the new data.

[0062] S2. Trigger the image information in the combined cache library to perform imaging according to preset imaging rules, obtain a two-dimensional plane map, and generate a road section unique code identifier for the two-dimensional plane image. The address information recorded in the frame stream image can be obtained through the road section unique code identifier.

[0063] Specifically, the preset imaging rule is to record the vehicle speed and time in real time after receiving the starting point trigger signal, and calculate the current distance value. When the current distance value reaches the preset distance threshold, the acquisition of the image information at the preset vertical coordinate position in the highway marking frame stream screen is stopped, and at the same time, the acquisition of the image information at the preset vertical coordinate position in the highway marking frame stream screen is restarted for the next section until the vehicle stop signal is received.

[0064] Among them, in this embodiment, the preset distance threshold can be set to 1 kilometer. When the vehicle travels 1 kilometer and ends the acquisition of image information, it also triggers the image information in the combined cache library to be imaged. Specifically, the pixel values ​​arranged in time sequence in the cache library are combined together to obtain a two-dimensional plane map, which is equivalent to converting the timing information of the center of the picture into plane information. In addition, a road section unique code identifier is generated for the two-dimensional plane map. The set road section unique code identifier is associated with the address information recorded in the frame stream picture. The setting rule of the road section unique code identifier is also set according to the starting address plus the distance value, so that the address information of the two-dimensional plane map can be identified by storing the road section unique code identifier and the starting address, effectively reducing the storage capacity of the device.

[0065] also, Figure 2 This is a flow chart of an optional implementation method for imaging the image information in the combined cache library in step S2 of the present disclosure, referring to Figure 2 , step S2 specifically includes the following steps S20, S21 and S22:

[0066] S20, matching pixel values ​​in two adjacent columns recorded in the cache to obtain similarity.

[0067] S21. When the similarity is greater than a preset threshold, similar contents are deleted according to a preset deletion rule.

[0068] S22. Combining the remaining pixel values ​​in the cache according to the chronological order to obtain a two-dimensional plane image.

[0069] Specifically, since the frame rate of a frame stream image is often faster than the driving speed of the vehicle, it is easy to take frames multiple times at the same position. If the pixel values ​​in the cache are directly combined, the actual marking line will be stretched too long, resulting in data redundancy that affects the overall processing speed. Therefore, similar content needs to be deleted. In this embodiment, the cosine similarity algorithm is preferably used to calculate the similarity of two adjacent columns of pixel values ​​in the cache. Cosine similarity describes the similarity of changes in two columns of pixels. The result is accurate but the amount of calculation is large. Therefore, the cosine similarity algorithm is very suitable for similar vector calculations with high precision requirements but limited number. According to the preset imaging rules, it can be known that the cache has a fixed length. Therefore, the cosine similarity algorithm is very suitable for the similarity calculation of this application. The cosine similarity calculation formula is as follows, where A and B represent two adjacent rows of pixel values, respectively, A·B represents the inner product of two adjacent rows of pixel values, and ‖A‖*‖B‖ represents the multiplication of the norms of two adjacent rows of pixel values:

[0070]

[0071] When the cosine similarity exceeds 95%, it is determined that the two adjacent rows of pixel values ​​are similar, and only the row of pixel values ​​with the later time is retained; otherwise, it is considered that the two adjacent rows of pixel values ​​are not similar, and both are retained. The above process is repeated until the entire cache is iterated once.

[0072] In another embodiment, the similarity between two adjacent rows of pixel values ​​in the calculation buffer may be obtained by performing a difference comparison between the two adjacent rows of pixel values.

[0073] S3. Extract and calculate the two-dimensional plane image to obtain a marked rectangular image.

[0074] Specifically, after the pixel values ​​in the cache are combined, the combined size of each section of the road is different, which can easily cause errors in the subsequent qualified judgment. Therefore, it is also necessary to perform relevant feature extraction, calculation and other processing on the two-dimensional plane map to obtain a uniform-sized marked rectangular image.

[0075] Figure 3 This is a flow chart of an optional implementation method of step S3 in the present disclosure, refer to Figure 3 , step S3 specifically includes the following steps S30, S31, S32 and S33:

[0076] S30, preprocessing the two-dimensional plane image to obtain an enhanced grayscale image.

[0077] S31, performing edge detection on the enhanced grayscale image to obtain marking line edge information.

[0078] S32, performing Hough transform on the marking line edge information to identify the marking line shape, and extracting the marking line shape to obtain a marking line image.

[0079] S33, pasting the marking line image into a layer of a preset fixed size, and automatically filling the preset pixel value into the blank space of the layer to obtain a marking line rectangular image.

[0080] Specifically, due to the influence of illumination, errors are likely to exist between the two-dimensional plane images of each section of the route. Therefore, it is necessary to enhance the two-dimensional plane image preprocessing to obtain an enhanced grayscale image. The enhanced grayscale image is edge detected to obtain the edge information of the marking, that is, the coordinates of several edge points of the marking. Among them, the edge detection can use Canny detection. Canny detection can identify as many actual edges as possible, while reducing the false alarms caused by noise as much as possible, and has the characteristics of high positioning and minimum response. Then, Hough transform is performed according to the edge information of the marking to identify the shape of the marking, and the shape of the marking is cut out to obtain the marking image. The marking image is then pasted into a layer of a preset fixed size, and the preset pixel value is automatically filled into the blank space of the layer to obtain a rectangular image of the marking, wherein the preset pixel value is set to 0, forming the maximum variance with the marking, which is conducive to eliminating the influence of the road surface other than the marking in the original image on the recognition of the marking quality.

[0081] Figure 4 This is a flowchart of an optional implementation method of step S30 in the present disclosure, refer to Figure 4 , step S30 specifically includes the following steps S300, S301 and S302:

[0082] S300, performing an opening operation of first corroding and then dilating the two-dimensional plane image to obtain a primary processed image.

[0083] S301, graying the primary processed image by weighted average method to obtain a gray image.

[0084] S302: Perform median filtering on the grayscale image to obtain an enhanced grayscale image.

[0085] Specifically, the two-dimensional plane image is first eroded and then expanded to obtain a primary processed image, which has the effect of eliminating small objects, separating objects at thin places and smoothing the boundaries of larger objects. Then the primary processed image is grayed. In this embodiment, the gray processing adopts the weighted average method. Then the gray image is subjected to median filtering to obtain an enhanced gray image. The median filter has a significant effect on salt and pepper noise and speckle noise, and the median filter has a good edge retention characteristic, which can further enhance the distinction of the situation where the edge of the marking line is incomplete.

[0086] S4. Determine whether the road marking rectangular image is qualified according to a preset judgment rule. If it is unqualified, retain the road section unique code identifier in the database.

[0087] Specifically, the preset judgment rule is to count the number of pixels in the standard rectangular image that are less than the preset threshold, and judge whether the ratio is greater than the preset proportion value based on the ratio of the calculated number to the total number. If the ratio is greater than the preset proportion value, the marking rectangular image is judged to be unqualified. After the failure, the unique code of the road section is saved, and the data in the cache is cleared to release the memory in time, enhance the sustainability of the recognition operation, and at the same time, the device cache capacity does not need to be too large, which greatly reduces the cost of equipment purchase.

[0088] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0089] The embodiment of the present application also provides an intelligent highway maintenance marking line quality detection system, which corresponds one-to-one with the intelligent highway maintenance marking line quality detection method in the embodiment. Figure 5 The intelligent highway maintenance marking quality detection system includes: an information recording module 1, an information imaging module 2, an image processing module 3 and an image judgment module 4. The detailed description of each functional module is as follows:

[0090] The information recording module 1 is used to obtain the image information at a preset vertical coordinate position in the highway marking frame stream according to the starting point trigger signal, and record the image information in a cache library.

[0091] The information imaging module 2 is used to trigger the image information in the combined cache library to be imaged according to the preset imaging rules, obtain a two-dimensional plane map, and generate a road section unique code identification for the two-dimensional plane image. The address information recorded in the frame stream image can be obtained through the road section unique code identification.

[0092] The image processing module 3 is used to extract and calculate the two-dimensional plane image to obtain a marked line rectangular image.

[0093] The image judgment module 4 is used to judge whether the road marking rectangular image is qualified according to a preset judgment rule. If it is unqualified, the road section unique code identification is retained in the database.

[0094] Among them, the intelligent highway maintenance marking quality detection system uses the information recording module 1 to obtain the image information at the preset vertical coordinate position in the highway marking frame stream image from the camera according to the starting point trigger signal, and records the image information in the cache library, and then uses the information imaging module 2 to trigger the image information in the combined cache library for imaging according to the preset imaging rules to obtain a two-dimensional plane map, and generate a road section unique code identification for the two-dimensional plane image. The address information recorded in the frame stream image can be obtained through the road section unique code identification, and then the image processing module 3 is used to extract and calculate the two-dimensional plane map to obtain the marking rectangular image, and finally the image judgment module 4 is used to judge whether the marking rectangular image is qualified according to the preset judgment rules. If it is unqualified, the road section unique code identification is retained in the database.

[0095] For the specific definition of the intelligent highway maintenance line marking quality detection system, please refer to the definition of the intelligent highway maintenance line marking quality detection method in the context, which will not be repeated here. Each module in the above-mentioned intelligent highway maintenance line marking quality detection system can be implemented in whole or in part by software, hardware and their combination. The above-mentioned modules can be embedded in or independent of the processor in the electronic device in the form of hardware, or can be stored in the memory of the electronic device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0096] The embodiment of the present application discloses an electronic device. An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, the following steps are performed:

[0097] S1. Obtain image information at a preset vertical coordinate position in a highway marking frame stream according to a starting point trigger signal, and record the image information in a cache library.

[0098] S2. Trigger the image information in the combined cache library to perform imaging according to preset imaging rules, obtain a two-dimensional plane map, and generate a road section unique code identifier for the two-dimensional plane image. The address information recorded in the frame stream image can be obtained through the road section unique code identifier.

[0099] S3. Extract and calculate the two-dimensional plane image to obtain a marked rectangular image.

[0100] S4. Determine whether the road marking rectangular image is qualified according to a preset judgment rule. If it is unqualified, retain the road section unique code identifier in the database.

[0101] The step of combining the image information in the cache library to perform imaging in step S2 includes:

[0102] S20, matching pixel values ​​in two adjacent columns recorded in the cache to obtain similarity.

[0103] S21. When the similarity is greater than a preset threshold, similar contents are deleted according to a preset deletion rule.

[0104] S22. Combining the remaining pixel values ​​in the cache according to the chronological order to obtain a two-dimensional plane image.

[0105] Step S3 includes:

[0106] S30, preprocessing the two-dimensional plane image to obtain an enhanced grayscale image.

[0107] S31, performing edge detection on the enhanced grayscale image to obtain marking line edge information.

[0108] S32, performing Hough transform on the marking line edge information to identify the marking line shape, and extracting the marking line shape to obtain a marking line image.

[0109] S33, pasting the marking line image into a layer of a preset fixed size, and automatically filling the preset pixel value into the blank space of the layer to obtain a marking line rectangular image.

[0110] Step S30 includes:

[0111] S300, performing an opening operation of first corroding and then dilating the two-dimensional plane image to obtain a primary processed image.

[0112] S301, graying the primary processed image by weighted average method to obtain a gray image.

[0113] S302: Perform median filtering on the grayscale image to obtain an enhanced grayscale image.

[0114] An embodiment of the present application also discloses a computer-readable storage medium storing a computer program that can be loaded and executed by a processor. When the computer program is executed by the processor, the steps of any of the above-mentioned intelligent highway maintenance marking quality detection methods are implemented and the same effect can be achieved.

[0115] The computer-readable storage medium includes, for example, a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0116] The above are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Any feature disclosed in this specification (including the abstract and drawings), unless otherwise stated, can be replaced by other equivalent or alternative features with similar purposes. That is, unless otherwise stated, each feature is only an example of a series of equivalent or similar features.

Claims

1. An intelligent highway maintenance marking quality detection method, characterized in that: include: According to the starting point trigger signal, the image information at the preset vertical coordinate position in the highway marking frame stream screen is obtained, that is, the pixel value in the horizontal direction of the most middle area of ​​the vertical coordinate position of the frame stream screen is obtained, and the image information is recorded in the cache library; Triggering the combination of image information in the cache library to perform imaging according to preset imaging rules to obtain a two-dimensional plane map, including: matching pixel values ​​recorded in two adjacent columns in the cache library to obtain similarity; when the similarity is greater than a preset threshold, deleting similar content according to a preset deletion rule; combining the remaining pixel values ​​in the cache library according to the time sequence to obtain a two-dimensional plane map; generating a road section unique code identifier for the two-dimensional plane image, and obtaining the address information recorded in the frame stream screen through the road section unique code identifier; Extract and calculate the two-dimensional plane image to obtain a marked line rectangular image; The road marking rectangular image is judged to be qualified according to the preset judgment rules. If it is unqualified, the road section unique code is retained in the database.

2. The intelligent highway maintenance line marking quality detection method according to claim 1 is characterized in that: The preset imaging rules include: After receiving the starting point trigger signal, the vehicle speed and time are recorded in real time, and the current distance value is calculated. When the current distance value reaches the preset distance threshold, the acquisition of the image information at the preset vertical coordinate position in the highway marking frame stream screen is stopped, and at the same time, the acquisition of the image information at the preset vertical coordinate position in the highway marking frame stream screen is restarted until the vehicle stop signal is received.

3. The intelligent highway maintenance line marking quality detection method according to claim 2 is characterized in that: In the step of matching the pixel values ​​recorded in two adjacent columns in the cache to obtain the similarity, cosine similarity is used to calculate the similarity of the pixel values ​​of two adjacent rows in the cache. The cosine similarity algorithm is as follows: Among them, A and B represent the pixel values ​​of two adjacent rows respectively, and A·B represents the inner product of the pixel values ​​of two rows. ‖A‖*‖B‖ represents the multiplication of the norms of two rows of pixel values.

4. The intelligent highway maintenance line marking quality detection method according to claim 1 is characterized in that: The step of extracting and calculating the two-dimensional plane image to obtain the marking line image includes: Preprocess the two-dimensional plane image to obtain an enhanced grayscale image; Perform edge detection on the enhanced grayscale image to obtain the edge information of the marking line; Performing Hough transform on the edge information of the marking line to identify the shape of the marking line, and extracting the shape of the marking line to obtain a marking line image; Paste the marking line image into a layer of preset fixed size, and automatically fill the preset pixel value into the blank space of the layer to obtain the marking line rectangular image.

5. The intelligent highway maintenance line marking quality detection method according to claim 4 is characterized in that: The step of preprocessing the two-dimensional plane image to obtain an enhanced grayscale image comprises: The two-dimensional plane image is opened by first corroding and then dilating to obtain a primary processed image; the primary processed image is grayed by weighted averaging to obtain a grayscale image; the grayscale image is median filtered to obtain an enhanced grayscale image.

6. The intelligent highway maintenance line marking quality detection method according to claim 1 is characterized in that: The preset judgment rules include: The number of pixel values ​​in the standard rectangular image that are less than a preset threshold is counted, and based on the ratio of the number to the total number, it is determined whether the ratio is greater than a preset ratio. If the ratio is greater than the preset ratio, the marked rectangular image is determined to be unqualified.

7. An intelligent highway maintenance marking quality detection system, characterized in that: include: An information recording module (1) is used to obtain image information at a preset vertical coordinate position in a highway marking frame stream image according to a starting point trigger signal, that is, to obtain the pixel value in the horizontal direction of the most central area of ​​the vertical coordinate position of the frame stream image, and record the image information in a cache library; The information imaging module (2) is used to trigger the combination of image information in the cache library to perform imaging according to a preset imaging rule to obtain a two-dimensional plane map, including: matching pixel values ​​recorded in two adjacent columns in the cache library to obtain similarity; when the similarity is greater than a preset threshold, deleting similar content according to a preset deletion rule; combining the remaining pixel values ​​in the cache library according to a time sequence to obtain a two-dimensional plane map; generating a road section unique code identifier for the two-dimensional plane image, and obtaining address information recorded in the frame stream image through the road section unique code identifier; An image processing module (3) is used to extract and calculate the two-dimensional plane image to obtain a marked rectangular image; The image judgment module (4) is used to judge whether the road marking rectangular image is qualified according to a preset judgment rule, and if it is unqualified, the road section unique code identifier is retained in the database.

8. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executes any one of the intelligent highway maintenance marking quality detection methods as claimed in claims 1 to 6.

9. A computer-readable storage medium, characterized in that: A computer program capable of being loaded by a processor and executing any one of claims 1 to 6 of an intelligent highway maintenance line marking quality detection method is stored.

Images