An automatic leveling and compacting device for road foundation

Abstract

The present invention relates to the technical field of image edge analysis, and in particular to an automatic leveling and compacting device for a road foundation. First, according to the characteristic that an area of ​​a road foundation that is not completely compacted and leveled usually shows certain edge features in an image, all suspected uneven areas are firstly screened out according to the edge distribution in a grayscale image of the road foundation; then, according to the characteristic that the grayscale values ​​of the edges on both sides of a soil block generated when the road is uneven usually have a large difference under the influence of light, the brightness difference is calculated, thereby preliminarily screening out a preferred uneven area; then, considering that the edge of the soil block is usually not smooth enough, the possibility of an interference area is determined in combination with the uniform distribution of the gradient amplitude of the edge pixel points of the preferred uneven area and the stability of the edge change; thereby, a more accurate compacted uneven area is screened out according to the possibility of the interference area, so that the effect of performing a road foundation leveling and compacting treatment according to the compacted uneven area is better.

Description

Technical Field

[0001] The invention relates to the technical field of image edge analysis, and in particular to an automatic leveling and compacting device for a road foundation. Background Art

[0002] The devices used to compact and level the road foundation mainly include rollers, graders and hydraulic compactors. Rollers improve the compactness and bearing capacity of the foundation by rolling repeatedly, reducing settlement and deformation. Graders are responsible for leveling the surface of the foundation to ensure uniform construction quality. Hydraulic compactors have relatively large impact force and are suitable for small areas and corners where compaction requirements are high. Therefore, they are usually used in the last stage of the road foundation compaction process. Through the effective coordination of these devices, a solid and flat foundation can be provided, the performance of the road can be improved, settlement problems can be reduced, and the stability and driving comfort of the road can be enhanced.

[0003] In order to ensure the quality of the flattening and compaction of the road foundation, after the road foundation is compacted, that is, after the hydraulic compactor compacts the road foundation, it is usually necessary to perform a flatness detection on the compacted area to determine whether the flatness of the compacted area meets the requirements. Considering that the labor cost and time cost of the manual method of detecting the flatness of the road foundation are high, the existing technology usually uses image processing to perform edge detection on the grayscale image of the road foundation based on the characteristics that the uncompacted area usually has a height difference resulting in certain edge features, and the detected edge connected domain is regarded as an uneven area that has not been compacted.

[0004] However, it is necessary to consider that when there are many types of materials used for the road foundation, materials of different colors will interfere with the edge detection results, resulting in the detected edge connected domain may contain compacted flat areas; therefore, the accuracy of the compacted uneven areas obtained by the edge detection method in the prior art is low, resulting in poor effect of road foundation leveling and compaction based on the compacted uneven areas. Summary of the invention

[0005] The present application provides an automatic leveling and compacting device for a road foundation. The device first selects all suspected uneven areas according to the edge distribution in the grayscale image of the road foundation based on the characteristic that areas where the road foundation is not completely compacted and leveled usually show certain edge features in the image; then calculates the brightness difference according to the characteristic that the grayscale values ​​of the edges on both sides of the soil blocks generated when the road is uneven usually have a large difference under the influence of light, thereby preliminarily selecting the preferred uneven areas; then, considering the characteristic that the edges of the soil blocks are usually not smooth enough, the possibility of the interference area is determined in combination with the uniform distribution of the gradient amplitude of the edge pixels of the preferred uneven area and the stability of the edge changes; thereby selecting a more accurate compacted uneven area according to the possibility of the interference area, solving the problem of low accuracy of the compacted uneven area obtained by the edge detection method in the prior art, and achieving a better effect of the road foundation leveling and compacting treatment according to the compacted uneven area.

[0006] The present application provides an automatic leveling and compacting device for a road foundation, the device comprising:

[0007] The data acquisition preprocessing module is used for data processing. During the road foundation compaction process, the image acquisition device on the hydraulic compactor collects the road foundation grayscale image corresponding to each road foundation area after compaction; according to the edge distribution in the road foundation grayscale image, all suspected uneven areas are screened out;

[0008] A preferred uneven area screening module is used to determine the brightness difference of each suspected uneven area according to the relative deviation of the gray value distribution of the edge pixel points of each suspected uneven area; and screen out the preferred uneven area among all the suspected uneven areas according to the brightness difference;

[0009] The compacted uneven area screening module is used to determine the edge smoothness of each preferred uneven area according to the uniform distribution of the gradient amplitude of each edge pixel point in each preferred uneven area and the stability of the edge change; determine the possibility of the interference area of ​​each preferred uneven area according to the brightness difference and the edge smoothness of each preferred uneven area; and screen out all the compacted uneven areas in the road foundation grayscale image according to the possibility of the interference area;

[0010] The road foundation leveling and compacting processing module is used to perform road foundation leveling and compacting processing according to the uneven areas to be compacted.

[0011] Furthermore, the process of obtaining the suspected uneven area includes:

[0012] Performing canny edge detection on the road foundation grayscale image to obtain a corresponding road foundation edge image; taking each edge connected domain in the road foundation edge image as a suspected uneven area.

[0013] Furthermore, the process of obtaining the brightness difference includes:

[0014] The average grayscale value of all edge pixels in each suspected uneven area is taken as the edge standard grayscale value;

[0015] In each suspected uneven area, the difference between the grayscale value of each edge pixel and the edge standard grayscale value is used as the grayscale standard deviation of each edge pixel;

[0016] The mean of the grayscale standard deviation of all edge pixels in each suspected uneven region is normalized to determine the brightness difference of each suspected uneven region.

[0017] Furthermore, the process of obtaining the optimal uneven area includes:

[0018] The suspected uneven area corresponding to the brightness difference greater than the preset difference threshold is taken as the preferred uneven area.

[0019] Furthermore, the process of obtaining the edge smoothness includes:

[0020] Determine the normalized chain code of each preferred uneven region by the chain code method; take the average of all chain code values ​​in the first-order differential chain code corresponding to the normalized chain code as the edge change complexity of each preferred uneven region;

[0021] Determining the degree of edge gradient disorder of each preferred uneven region according to the disorder of grayscale gradient amplitude distribution corresponding to edge pixel points in each preferred uneven region;

[0022] The edge smoothness of each preferred uneven region is determined according to the edge variation complexity and the edge gradient disorder degree; the edge variation complexity and the edge gradient disorder degree are both negatively correlated with the edge smoothness.

[0023] Furthermore, the process of obtaining the edge gradient disorder degree includes:

[0024] The grayscale gradient amplitude corresponding to each edge pixel point in each preferred uneven region is counted; and the standard deviation of the grayscale gradient amplitudes of all edge pixel points corresponding to each preferred uneven region is used as the edge gradient disorder degree of each preferred uneven region.

[0025] Furthermore, the process of determining the edge smoothness of each preferred uneven region according to the edge variation complexity and the edge gradient disorder degree includes:

[0026] The product of the edge variation complexity and the edge gradient disorder degree is negatively correlated and mapped to determine the edge smoothness of each preferred uneven region.

[0027] Furthermore, the process of obtaining the possibility of the interference area includes:

[0028] The product of the negative correlation mapping value of the brightness difference of each preferred uneven region and the edge smoothness degree is normalized to determine the interference region possibility of each preferred uneven region.

[0029] Furthermore, the process of obtaining the compacted uneven area includes:

[0030] All preferred uneven areas corresponding to the interference area probability less than the preset interference threshold are regarded as compacted uneven areas.

[0031] Furthermore, the road foundation leveling and compacting process according to the compacted uneven area includes:

[0032] The road foundation area corresponding to the road foundation grayscale image containing the compacted uneven area is used as the area to be leveled and compacted for the second time; after the compaction of all road foundation areas is completed, secondary road foundation compaction is performed on all areas to be leveled and compacted for the second time.

[0033] This application includes the following benefits:

[0034] First, when the road foundation is not completely compacted and leveled, there will usually be protruding or redundant soil blocks, which usually have complete edges in the road foundation grayscale image. Therefore, firstly, all suspected uneven areas are screened out according to the edge distribution in the road foundation grayscale image.

[0035] Further considering that materials of different colors in the road foundation will also form edge connected domains in the road foundation grayscale image, that is, interference connected domains that interfere with the screening of compacted uneven areas, the existence of these interference connected domains results in that the screened suspected uneven areas are not all areas corresponding to protruding or redundant soil blocks that are usually not completely compacted and leveled, so it is necessary to conduct further analysis based on the suspected uneven areas. For the soil blocks that appear on the road foundation, there will be shadows on one side of the edge of the soil block under the influence of light, which is manifested in the image as a lower grayscale value on one side of the edge and a higher grayscale value on the other side of the edge; while the edge connected domains corresponding to the interfering road foundation materials are usually not affected by light, and the grayscale values ​​of the corresponding edge pixels are usually close; therefore, firstly, the brightness difference of each suspected uneven area is determined according to the relative deviation of the grayscale value distribution of the edge pixels of each suspected uneven area; thus, the preferred uneven areas among all suspected uneven areas are screened out according to the brightness difference.

[0036] Further considering that the road foundation is composed of a variety of materials, this may cause some interference connected domains to have large differences in grayscale values ​​on both sides of the edge, so further analysis is needed based on the preferred uneven area. For the connected domain corresponding to the soil block, in addition to the large grayscale difference between the edge pixels on both sides, its edge shape and grayscale change are usually not smooth, while the interference connected domain is compacted and flat and usually integrates with the road foundation more stably, so the corresponding edge shape and grayscale change are relatively smoother; therefore, the edge smoothness of each preferred uneven area is determined based on the uniform distribution of the gradient amplitude of each edge pixel in each preferred uneven area and the stability of the edge change; thus, based on the edge smoothness, a comprehensive analysis is further conducted in combination with the brightness difference of the preferred uneven area to screen out more accurate compacted uneven areas in the road foundation grayscale image; so that the subsequent road foundation leveling and compaction treatment based on the compacted uneven area is more effective. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0038] Figure 1 A structural block diagram of an automatic leveling and compacting device for a road foundation provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0039] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation method, structure, characteristics and effects of an automatic leveling and compacting device for road foundation proposed by the present invention are described in detail below in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment, and the specific features, structures or characteristics in one or more embodiments may be combined in any suitable form. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as implying or suggesting relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features.

[0040] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0041] The specific scheme of the automatic leveling and compacting device for road foundation provided by the present invention is described in detail below with reference to the accompanying drawings.

[0042] The present application provides an automatic leveling and compacting device for road foundation. Figure 1 , which shows a structural block diagram of an automatic leveling and compacting device for a road foundation provided by an embodiment of the present invention, including a data acquisition preprocessing module 101, an uneven area selection module 102, a compacting uneven area selection module 103 and a road foundation leveling and compacting processing module 104.

[0043] The data acquisition preprocessing module 101 is used for data processing. During the road foundation compaction process, the image acquisition device on the hydraulic compactor collects the road foundation grayscale image corresponding to each road foundation area after compaction; according to the edge distribution in the road foundation grayscale image, all suspected uneven areas are screened out.

[0044] In a specific implementation of an embodiment of the present invention, a high-precision industrial camera is installed behind a hydraulic compactor so that the road foundation compacted and leveled by the hydraulic compactor can be captured by the high-precision industrial camera, and in order to capture as many road foundation areas as possible; the height of the high-precision industrial camera is set to 1.5 meters, the shooting direction is vertically downward, and the shooting frequency is set to capture once per second, so that all the road foundation grayscale images captured can cover all road foundation areas, and each road foundation grayscale image obtained each time corresponds to a road foundation area. It should be noted that the height and shooting frequency of the high-precision industrial camera can be adjusted according to the specific implementation environment, and will not be further elaborated here.

[0045] First, when the road foundation is not completely compacted and leveled, there will usually be protruding or redundant soil blocks. These soil blocks usually have complete edges in the road foundation grayscale image. Therefore, all suspected uneven areas are first screened out according to the edge distribution in the road foundation grayscale image.

[0046] Preferably, in some possible implementations of the embodiments of the present invention, the process of acquiring the suspected uneven area includes:

[0047] Canny edge detection is performed on the road foundation grayscale image to obtain the corresponding road foundation edge image; each edge connected domain in the road foundation edge image is regarded as a suspected uneven area; the edge connected domain is an area with closed edges, and the edges of the compacted uneven area corresponding to the soil blocks generated during uneven compaction are usually relatively complete, so the compacted uneven area can be screened based on the suspected uneven area. It should be noted that the canny edge detection method is a technical means well known to those skilled in the art and will not be further described here.

[0048] The preferred uneven region screening module 102 is used to determine the brightness difference of each suspected uneven region according to the relative deviation of the gray value distribution of the edge pixels of each suspected uneven region; and screen out the preferred uneven region among all the suspected uneven regions according to the brightness difference.

[0049] Further consideration is given that materials of different colors in the road foundation will also form edge connected domains in the road foundation grayscale image, that is, interfering connected domains that will interfere with the screening of compacted uneven areas. The existence of these interfering connected domains results in that the screened suspected uneven areas are not all areas corresponding to protruding or redundant soil blocks that usually appear when the area is not completely compacted and leveled. Therefore, further analysis is required based on the suspected uneven areas. For the soil blocks appearing on the road foundation, there will be a shadow on one side of the edge of the soil block under the influence of light, which is manifested in the image as a lower grayscale value on one side of the edge and a higher grayscale value on the other side of the edge; the edge connection domain corresponding to the interfering road foundation material is usually not affected by light, and the grayscale values ​​of the corresponding edge pixels are usually close; therefore, firstly, the brightness difference of each suspected uneven area is determined according to the relative deviation of the grayscale value distribution of the edge pixels of each suspected uneven area; the larger the corresponding brightness difference, the more it conforms to the lighting characteristics corresponding to the soil block, that is, the more likely it is to belong to a real compacted uneven area; thus, the preferred uneven area among all suspected uneven areas is screened out according to the brightness difference.

[0050] Preferably, in some possible implementations of the embodiments of the present invention, the process of acquiring the brightness difference includes:

[0051] The mean grayscale value of all edge pixels in each suspected uneven area is taken as the edge standard grayscale value; in each suspected uneven area, the difference between the grayscale value of each edge pixel and the edge standard grayscale value is taken as the grayscale standard deviation of each edge pixel; the mean grayscale standard deviation of all edge pixels in each suspected uneven area is normalized to determine the brightness difference of each suspected uneven area. Since the real compacted uneven area corresponding to the soil block has a large grayscale difference between the edge pixels on both sides under the influence of light, the grayscale value of the edge standard grayscale value is usually between the overall grayscale values ​​on both sides, resulting in a large standard deviation of the grayscale values ​​of all edge pixels; that is, the greater the corresponding brightness difference, the greater the influence of light on the corresponding suspected uneven area, and the greater the possibility that the corresponding area belongs to the real compacted uneven area.

[0052] In a specific implementation of the embodiment of the present invention, the process of obtaining the brightness difference is expressed by the formula: ;in, is the first grayscale image of the road foundation The brightness difference of the suspected uneven area; is the first grayscale image of the road foundation The number of edge pixels in the suspected uneven area; is the first grayscale image of the road foundation In the suspected uneven area Gray value of edge pixels; is the first grayscale image of the road foundation In the suspected uneven area Gray value of edge pixels; is the first grayscale image of the road foundation The average gray value of all edge pixels in the suspected uneven area, that is, the standard gray value of the edge; is the first grayscale image of the road foundation In the suspected uneven area Grayscale standard deviation of edge pixels; It is a linear normalization function. The implementer can adjust the normalization method according to the specific implementation environment. Through normalization, the value range of brightness difference is limited to 0 to 1, which is convenient for subsequent selection of uneven areas.

[0053] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the uneven area preferably includes:

[0054] The suspected uneven area corresponding to the brightness difference greater than the preset difference threshold is used as the preferred uneven area. Since the greater the brightness difference, the more likely the corresponding suspected uneven area is to be the real compacted uneven area corresponding to the soil block, the preferred uneven area screening is performed by setting the threshold, thereby further narrowing the screening range of the compacted uneven area and reducing the complexity of subsequent analysis to a certain extent. In a specific implementation of an embodiment of the present invention, the preset difference threshold is set to 0.7, which can be adjusted according to the specific implementation environment.

[0055] The compacted uneven area screening module 103 is used to determine the edge smoothness of each preferred uneven area according to the uniformity of the gradient amplitude distribution of each edge pixel point in each preferred uneven area and the stability of the edge change; determine the possibility of the interference area of ​​each preferred uneven area according to the brightness difference and edge smoothness of each preferred uneven area; and screen out all the compacted uneven areas in the road foundation grayscale image according to the possibility of the interference area.

[0056] Further considering that the road foundation is composed of a variety of materials, this may cause some interference connected domains to have large differences in grayscale values ​​on both sides of the edge, so further analysis is needed based on the preferred uneven area. For the connected domain corresponding to the soil block, in addition to the large grayscale difference between the edge pixels on both sides, its edge shape is usually not smooth, while the interference connected domain is compacted and flat and usually integrates with the road foundation more stably, so the corresponding edge shape and grayscale gradient are relatively smoother; therefore, according to the uniform distribution of the gradient amplitude of each edge pixel in each preferred uneven area and the stability of the edge change, the edge smoothness of each preferred uneven area is determined, so that the greater the edge smoothness, the greater the possibility of belonging to the interference area, that is, the smaller the possibility of belonging to the real compacted uneven area corresponding to the soil block; further, based on the edge smoothness, the brightness difference of the preferred uneven area is further combined for comprehensive analysis to screen out more accurate compacted uneven areas in the road foundation grayscale image.

[0057] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the edge smoothness includes:

[0058] The normalized chain code of each preferred uneven area is determined by the chain code method; the mean of all chain code values ​​in the first-order differential chain code corresponding to the normalized chain code is used as the complexity of the edge change of each preferred uneven area. The size of each value in the first-order differential chain code of the normalized chain code characterizes the amplitude of the edge direction change to a certain extent. Therefore, the larger the overall mean of all chain code values ​​in the corresponding differential chain code, the less smooth the edge direction change is, and the greater the possibility that the corresponding uneven area belongs to an irregular soil block; that is, the greater the complexity of the edge change, the greater the possibility that the corresponding preferred uneven area belongs to the real compacted uneven area corresponding to the soil block; it should be noted that the chain code method is a technical means well known to those skilled in the art, and will not be further elaborated here.

[0059] For the real compacted uneven area corresponding to the soil block, the grayscale gradient amplitude of the corresponding edge pixel point is usually chaotic because the edge of the soil block is not smooth and is affected by light. For the interference area, different materials will be integrated together during the compaction process of the road foundation, and the corresponding grayscale changes are usually stable. Therefore, according to the chaotic distribution of the grayscale gradient amplitude corresponding to the edge pixel points in each preferred uneven area, the degree of edge gradient chaos of each preferred uneven area is determined, so that the greater the degree of edge gradient chaos, the less likely it is to belong to the interference area, and the greater the possibility of belonging to the real compacted flat area corresponding to the soil block.

[0060] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the degree of edge gradient disorder includes:

[0061] The grayscale gradient amplitude corresponding to each edge pixel in each preferred uneven region is counted; the standard deviation of the grayscale gradient amplitudes of all edge pixels corresponding to each preferred uneven region is used as the edge gradient disorder degree of each preferred uneven region. The standard deviation can characterize the discrete degree of a set of data, so the larger the standard deviation of the grayscale gradient amplitudes of all edge pixels corresponding to each preferred uneven region, the more discrete the numerical distribution of the grayscale gradient amplitudes of all corresponding edge pixels is, and the greater the corresponding edge gradient disorder degree is.

[0062] Further, according to the complexity of edge change and the degree of edge gradient chaos, the edge smoothness of each preferred uneven area is determined, so that the greater the edge smoothness, the less it conforms to the uneven edge characteristics of the real compacted flat area corresponding to the soil block, that is, the smaller the possibility that the corresponding preferred uneven area belongs to the real compacted flat area corresponding to the soil block; because the greater the edge gradient chaos, the greater the edge gradient chaos, the greater the possibility that the corresponding preferred uneven area belongs to the real compacted flat area corresponding to the soil block, so the complexity of edge change and the degree of edge gradient chaos are negatively correlated with the edge smoothness. In a specific implementation of an embodiment of the present invention, the process of obtaining the edge smoothness is expressed by the formula: ;in, is the first grayscale image of the road foundation The degree of edge smoothness of the optimal uneven area; is the first grayscale image of the road foundation The mean of all chain code values ​​in the first-order difference chain code corresponding to the normalized chain code of the optimal uneven area, that is, the complexity of edge change; is the first grayscale image of the road foundation The standard deviation of the grayscale gradient amplitude of all edge pixels corresponding to the optimal uneven area, that is, the degree of edge gradient disorder; is a linear normalization function. The implementer can adjust the normalization method according to the specific implementation environment. Realize Negative correlation mapping processing.

[0063] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the possibility of the interference area includes:

[0064] Since the greater the brightness difference and the smaller the edge smoothness, the greater the possibility that the corresponding preferred uneven area belongs to the real compacted uneven area corresponding to the soil block, that is, the smaller the possibility of belonging to the interference area, the product of the negative correlation mapping value of the brightness difference of each preferred uneven area and the edge smoothness is further normalized to determine the interference area possibility of each preferred uneven area. In a specific implementation of the embodiment of the present invention, the process of obtaining the interference area possibility is expressed by the formula: ;in, is the first grayscale image of the road foundation Possibility of interference area for optimal uneven areas; is the first grayscale image of the road foundation The brightness difference of the optimal uneven area; is the first grayscale image of the road foundation The degree of edge smoothness of the optimal uneven area; It is a linear normalization function. The implementer can adjust the normalization method according to the specific implementation environment. Through normalization, the possible value range of the interference area is limited to 0 to 1, which is convenient for the subsequent screening of uneven areas. is an exponential function with a natural constant as the base. Implementers can adopt other negative correlation mapping methods according to the specific implementation environment, such as , and the reciprocal, where is a hyperbolic tangent function, which will not be further described here.

[0065] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the compacted uneven area includes:

[0066] The greater the possibility of the interference area, the more likely the corresponding preferred uneven area is the interference area, that is, the smaller the possibility of being a real compacted uneven area, so all preferred uneven areas corresponding to the interference area possibility less than the preset interference threshold are further taken as compacted uneven areas. In a specific implementation of the embodiment of the present invention, the preset interference threshold is set to 0.24, which can be adjusted according to the specific implementation environment.

[0067] The road foundation leveling and compacting processing module 104 is used to perform a road foundation leveling and compacting process according to the compacted uneven areas.

[0068] After obtaining the real compacted uneven area corresponding to the soil block, it is necessary to further perform a second leveling and compacting process on the road foundation area corresponding to each road foundation grayscale image containing the compacted uneven area to ensure the compaction quality of the road foundation. In a specific implementation of an embodiment of the present invention, performing a road foundation leveling and compacting process according to the compacted uneven area includes: taking the road foundation area corresponding to the road foundation grayscale image containing the compacted uneven area as the area to be leveled and compacted for the second time; after the compaction of all road foundation areas is completed, performing a second road foundation compaction on all areas to be leveled and compacted for the second time.

[0069] To summarize, the present application first selects all suspected uneven areas based on the edge distribution in the grayscale image of the road foundation based on the fact that areas where the road foundation has not been completely compacted and leveled usually show certain edge features in the image; then, based on the fact that the grayscale values ​​of the edges on both sides of the soil blocks generated when the road is uneven usually have large differences under the influence of light, the brightness difference is calculated, thereby preliminarily selecting the preferred uneven areas; then, taking into account the fact that the edges of the soil blocks are usually not smooth enough, the possibility of the interference area is determined in combination with the uniform distribution of the gradient amplitude of the edge pixels of the preferred uneven area and the stability of the edge changes; thereby, a more accurate compacted uneven area is selected based on the possibility of the interference area, thereby solving the problem of low accuracy of the compacted uneven area obtained by the edge detection method in the prior art, and achieving better effect of leveling and compacting the road foundation based on the compacted uneven area.

[0070] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0071] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. An automatic leveling and compacting device for road foundation, characterized in that: The device comprises: The data acquisition preprocessing module is used to collect the road foundation grayscale image corresponding to each road foundation area after compaction through the image acquisition device on the hydraulic compactor during the road foundation compaction process; and screen out all suspected uneven areas according to the edge distribution in the road foundation grayscale image; A preferred uneven area screening module is used to determine the brightness difference of each suspected uneven area according to the relative deviation of the gray value distribution of the edge pixel points of each suspected uneven area; and screen out the preferred uneven area among all the suspected uneven areas according to the brightness difference; The compacted uneven area screening module is used to determine the edge smoothness of each preferred uneven area according to the uniform distribution of the gradient amplitude of each edge pixel point in each preferred uneven area and the stability of the edge change; determine the possibility of the interference area of ​​each preferred uneven area according to the brightness difference and the edge smoothness of each preferred uneven area; and screen out all the compacted uneven areas in the road foundation grayscale image according to the possibility of the interference area; The road foundation leveling and compacting processing module is used to perform road foundation leveling and compacting processing according to the uneven areas to be compacted.

2. The automatic leveling and compacting device for road foundation according to claim 1, characterized in that: The process of obtaining the suspected uneven area includes: Performing canny edge detection on the road foundation grayscale image to obtain a corresponding road foundation edge image; taking each edge connected domain in the road foundation edge image as a suspected uneven area.

3. The automatic leveling and compacting device for road foundation according to claim 1, characterized in that: The process of obtaining the brightness difference includes: The average grayscale value of all edge pixels in each suspected uneven area is taken as the edge standard grayscale value; In each suspected uneven area, the difference between the grayscale value of each edge pixel and the edge standard grayscale value is used as the grayscale standard deviation of each edge pixel; The mean of the grayscale standard deviation of all edge pixels in each suspected uneven region is normalized to determine the brightness difference of each suspected uneven region.

4. The automatic leveling and compacting device for road foundation according to claim 1, characterized in that: The process of obtaining the preferred uneven area includes: The suspected uneven area corresponding to the brightness difference greater than the preset difference threshold is taken as the preferred uneven area.

5. The automatic leveling and compacting device for road foundation according to claim 1, characterized in that: The process of obtaining the edge smoothness includes: Determine the normalized chain code of each preferred uneven region by the chain code method; take the average of all chain code values ​​in the first-order differential chain code corresponding to the normalized chain code as the edge change complexity of each preferred uneven region; Determining the degree of edge gradient disorder of each preferred uneven region according to the disorder of grayscale gradient amplitude distribution corresponding to edge pixel points in each preferred uneven region; The edge smoothness of each preferred uneven region is determined according to the edge variation complexity and the edge gradient disorder degree; the edge variation complexity and the edge gradient disorder degree are both negatively correlated with the edge smoothness.

6. The automatic leveling and compacting device for road foundation according to claim 5, characterized in that: The process of obtaining the edge gradient disorder degree includes: The grayscale gradient amplitude corresponding to each edge pixel point in each preferred uneven region is counted; and the standard deviation of the grayscale gradient amplitudes of all edge pixel points corresponding to each preferred uneven region is used as the edge gradient disorder degree of each preferred uneven region.

7. The automatic leveling and compacting device for road foundation according to claim 5, characterized in that: The process of determining the edge smoothness of each preferred uneven region according to the edge variation complexity and the edge gradient disorder degree comprises: The product of the edge variation complexity and the edge gradient disorder degree is negatively correlated and mapped to determine the edge smoothness of each preferred uneven region.

8. The automatic leveling and compacting device for road foundation according to claim 1, characterized in that: The process of obtaining the possibility of the interference area includes: The product of the negative correlation mapping value of the brightness difference of each preferred uneven region and the edge smoothness degree is normalized to determine the interference region possibility of each preferred uneven region.

9. The automatic leveling and compacting device for road foundation according to claim 1, characterized in that: The acquisition process of compacting the uneven area includes: All preferred uneven areas corresponding to the interference area probability less than the preset interference threshold are regarded as compacted uneven areas.

10. The automatic leveling and compacting device for road foundation according to claim 1, characterized in that: The road foundation leveling and compacting process according to the compaction of the uneven area includes: The road foundation area corresponding to the road foundation grayscale image containing the compacted uneven area is used as the area to be leveled and compacted for the second time; after the compaction of all road foundation areas is completed, secondary road foundation compaction is performed on all areas to be leveled and compacted for the second time.

Images