A method for determining the particle size of roadbed filler using image processing
Through image processing technology, the efficiency and accuracy of the determination of the particle size of the roadbed filler is solved. Combined with the physical properties of the filler, the automatic measurement of the particle size of the roadbed is realized, and the operation efficiency and result reliability are improved.
Patent Information
- Application Number
- CN202411261338.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-10
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-09-10
AI Technical Summary
The prior art is difficult to accurately and efficiently determine the particle size of the roadbed filler, and fails to consider the impact of the particle size of the roadbed filler on the stability of the base cement and gravel cracks, resulting in low operating efficiency and poor economicality.
By using image processing technology, by collecting images of stable gravels in roadbed cement, crack shadow removal, main line extraction, length calculation, filler absorption cement coefficient calculation and particle size correction, it is automatically determined based on the apparent density of filler and sieve hole grading. Subgrade filler particle size.
It realizes accurate and rapid determination of the particle size of the roadbed filler, reduces the randomness of manual operations, improves the operation efficiency, and takes into account the cement stability and crack influence, which has high reliability and applicability.
Smart Images

Figure SMS_64 
Figure QLYQS_1 
Figure QLYQS_5
Abstract
Description
Technical Field
[0001] The present invention relates to the field of information model technology, and in particular to the application of image processing technology in roadbed engineering. The present invention proposes a method for determining the particle size of roadbed filler using image processing. Background Art
[0002] Roadbed filler is a material used to fill the roadbed. It is required to be easy to dig, easy to compact, high in strength and good in water stability. If the filler particle size is too large, it is easy to cause excessive porosity and the roadbed compaction will be problematic and unable to bear the load. If the filler particle size is too small, the crushing process is cumbersome and the economy is poor.
[0003] Generally, the maximum particle size of the filler is determined through soil sampling tests, but there are problems such as random soil sampling and low efficiency of manual operation, and the impact of the particle size of roadbed filler on the cracks of the base cement stabilized gravel is not considered. Therefore, a relatively accurate and efficient operation method is urgently needed to determine the particle size of roadbed filler. Summary of the invention
[0004] In order to solve the problem of how to determine the particle size of roadbed filler, the present invention proposes a method for determining the particle size of roadbed filler by using image processing.
[0005] In order to achieve the above object, the present invention provides the following scheme:
[0006] A method for determining the particle size of roadbed filler by image processing comprises the following steps:
[0007] Step 1, using a camera to collect images of cement concrete pavement constructed with cement stabilized gravel for roadbed;
[0008] Step 2, removing crack shadows from the image collected in step 1;
[0009] Step 3, extracting the main line type of the crack from the image after removing the shadow;
[0010] Step 4: Calculate the crack length;
[0011] Scan the crack image converted into the main line type pixel point. If the gray value of a pixel point is found to change from 255 to 0, then this pixel point is considered to be the uppermost pixel point of the crack. If the gray value of a pixel point is found to change from 0 to 255, it is considered to be the lowermost pixel point.
[0012] The crack length is calculated as follows:
[0013] ;
[0014] Where: —crack length in the image, m;
[0015] —The number of pixels in the image;
[0016] —The number of pixels in the image;
[0017] —The coordinates of any pixel of the crack in the image;
[0018] —The coordinates of another pixel of the crack in the image;
[0019] Step 5, the crack length and the composite apparent density of the filler are used to correct the specific surface area;
[0020] ;
[0021] Where: —Corrected value of filler specific surface area, ;
[0022] —Specific surface area of filler, ;
[0023] —Average sieve hole classification residual percentage of filler, ;
[0024] — synthetic wool volume density of the filler, ;
[0025] Step six, calculate the cement absorption coefficient of the filler;
[0026] ;
[0027] Where: — filler absorption coefficient of cement;
[0028] —Effective relative density of filler;
[0029] —composite apparent density of the filler, ;
[0030] - relative density of cement;
[0031] Step seven, calculating the roadbed filler particle size;
[0032] ;
[0033] Where: —Grain size of roadbed filler, ;
[0034] — filler absorption coefficient of cement;
[0035] After D is calculated, take the difference between D and the sieve hole sizes of 0.075, 0.15, 0.3, 0.6, 1.18, 2.36, 4.75, 9.5, 13.2, and 16. The sieve hole size corresponding to the smallest difference is the final roadbed filler particle size.
[0036] Step 2 is as follows:
[0037] (1) Match the structural element with the pixels in the image. If the structural element contains the image pixel, the image pixel value and the surrounding pixel values are enhanced. If the structural element does not contain the pixel, no processing is performed;
[0038] (2) After the operation in (1), the result is subjected to Gaussian filtering;
[0039] (3) Divide the shadow part by threshold segmentation, perform the maximum entropy threshold segmentation operation on the processing result of (2), and obtain the decision value of whether to segment, which is used as the segmentation threshold;
[0040] (4) Calculate the standard deviation of pixel brightness in the non-shadow area and the average pixel brightness , the road image is divided into N = 256 regions of different brightness, Make the area Contains brightness value All pixels of , ;
[0041] (5) Calculate the image pixel compensation coefficient;
[0042] ;
[0043] Where: —Image pixel compensation coefficient;
[0044] —The average pixel brightness value of the non-shadow part;
[0045] —The average pixel brightness value of the shadow part;
[0046] —The standard deviation of pixel brightness in the non-shadow part;
[0047] —The standard deviation of pixel brightness in the shadow part;
[0048] (6) Compensate the brightness value of the pixels in the image;
[0049] ;
[0050] Where: —The brightness value of each pixel of the processed road surface image;
[0051] —The brightness value of each pixel of the processed road surface image after compensation;
[0052] (7) Binarize the image.
[0053] Step three is as follows: initialize the binarized image, process each pixel one by one in a predetermined order, and decide whether to perform stripping based on the situation of the surrounding pixels. If at least one pixel around a pixel is not an edge pixel, the pixel cannot be stripped, otherwise the pixel can be stripped. In the segmented binary image, the grayscale value of the retained points is assigned to 0, and the grayscale values of the remaining points are assigned to 255.
[0054] The beneficial effects of the present invention are:
[0055] The present invention proposes an image processing method for determining the particle size of roadbed fillers, which creatively uses automatic image processing and model calculation methods to establish the relationship between the filler particle size and the cracks of the roadbed cement, thereby realizing the determination of the filler particle size. By removing the shadows of the collected images, the environmental factors on both sides of the road, such as the shadows of trees, are prevented from affecting the shooting and identification of road cracks. At the same time, the shadow part is divided by threshold segmentation to realize self-adjustment and determination of the shadow part division threshold. Brightness compensation is performed on the pixel points, so that the main line type of the cracks in the image can be better extracted, thereby realizing the accurate calculation of the crack length. By determining the total specific surface area of the filler, considering the effective cement content at the optimal cement film thickness and the amount of cement absorbed by the filler, the roadbed filler particle size is obtained, which has good reliability and applicability. DETAILED DESCRIPTION
[0056] In order to make the technical means, creative features, purpose and efficacy of the present invention easy to understand, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain this description and are not intended to limit the present invention.
[0057] Embodiment: A method for determining the particle size of roadbed filler by image processing, comprising the following steps:
[0058] Step 1, using a camera to collect images of a cement concrete pavement constructed with cement stabilized gravel for roadbed;
[0059] Step 2, removing crack shadows from the image collected in step 1;
[0060] (1) Match the structural element with the pixels in the image. If the structural element contains the image pixel, the image pixel value and the surrounding pixel values are enhanced, the structural element and the pixel in the image are overlapped, and the maximum pixel value contained in the structural element is taken as the new pixel value of the pixel; if the structural element does not contain the pixel, no processing is performed;
[0061] (2) After the operation in (1), the result is subjected to Gaussian filtering;
[0062] (3) Divide the shadow part by threshold segmentation. Perform the maximum entropy threshold segmentation operation on the processing result of (2) to obtain the decision value of whether to segment or not, which is used as the segmentation threshold;
[0063] (4) Calculate the standard deviation of pixel brightness in the non-shadow area and the average pixel brightness , the road image is divided into N = 256 regions of different brightness, Make the area Contains brightness value All pixels of , ;
[0064] (5) Calculate the image pixel compensation coefficient;
[0065] ;
[0066] Where: —Image pixel compensation coefficient;
[0067] —The average pixel brightness value of the non-shadow part;
[0068] —The average pixel brightness value of the shadow part;
[0069] —The standard deviation of pixel brightness in the non-shadow part;
[0070] —The standard deviation of pixel brightness in the shadow part;
[0071] (6) Compensate the brightness value of the pixels in the image;
[0072] ;
[0073] Where: —The brightness value of each pixel of the processed road surface image;
[0074] —The brightness value of each pixel of the processed road surface image after compensation;
[0075] (7) Binarize the image.
[0076] Step 3, extracting the main line type of the crack from the image after removing the shadow;
[0077] First, the binarized image is initialized, and then each pixel is processed one by one in a predetermined order, and whether stripping is required is determined based on the situation of the surrounding pixels. Specifically, if at least one pixel around a pixel is not an edge pixel, the pixel cannot be stripped, otherwise the pixel can be stripped; in the segmented binary image, the grayscale value of the retained point is assigned to 0, and the grayscale value of the remaining points is assigned to 255.
[0078] Step 4: Calculate the crack length;
[0079] Scan the crack image converted into the main line type pixel by pixel, starting from the upper left corner and scanning the pixels from top to bottom. If the gray value of a pixel is found to change from 255 to 0, then this pixel is considered to be the uppermost pixel of the crack. Conversely, if the gray value of a pixel is found to change from 0 to 255, it is considered to be the lowermost pixel.
[0080] The crack length is calculated as follows:
[0081] ;
[0082] Where: —crack length in the image, m;
[0083] —The number of pixels in the image;
[0084] —The number of pixels in the image;
[0085] —The coordinates of any pixel of the crack in the image;
[0086] —The coordinates of another pixel of the crack in the image.
[0087] Computer image processing software automatically calculated the crack length in the image to be 0.2 m.
[0088] Step 5, the crack length and the composite apparent density of the filler are used to correct the specific surface area;
[0089] The filler specific surface area calculation information table in this embodiment is as follows:
[0090]
[0091] Synthetic wool bulk density of filler is 2.755 , filler specific surface area 0.716 , the average sieve hole classification sieve residue percentage of the filler It is 12.5%.
[0092] Substitute the following formula into the calculation:
[0093] ;
[0094] Where: —Corrected value of filler specific surface area, ;
[0095] —Specific surface area of filler, ;
[0096] —Average sieve hole classification residual percentage of filler, ;
[0097] — synthetic wool volume density of the filler, ;
[0098] The calculated value is 0.956 .
[0099] Step 6, calculating the cement absorption coefficient of filler;
[0100] ;
[0101] Where: — filler absorption coefficient of cement, ;
[0102] —The effective relative density of the filler is 2.755;
[0103] —The composite apparent density of the filler is 2.753 ;
[0104] —The relative density of cement is 3.1.
[0105] Calculate the filler absorption cement coefficient It is 0.062.
[0106] Step 7, calculating the particle size of the roadbed filler;
[0107] ;
[0108] Where: —Grain size of roadbed filler, ;
[0109] — Filler absorption coefficient of cement, take 0.062;
[0110] The calculated particle size D of the roadbed filler is 7.82 mm, which has the smallest difference with the 9.5 mm sieve hole size, so 9.5 mm can be used as the final roadbed filler particle size.
[0111] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.
Claims
1. A method for determining the particle size of roadbed filler using image processing, characterized in that: The following steps are involved: Step 1, using a camera to collect images of cement concrete pavement constructed with cement stabilized gravel for roadbed; Step 2, removing crack shadows from the image collected in step 1; Step 3, extracting the main line type of the crack from the image after removing the shadow; Step 4: Calculate the crack length; Scan the crack image converted into the main line type pixel by pixel, and calculate the crack length according to the following formula: ; Where: —crack length in the image, m; —The number of pixels in the image; —The number of pixels in the image; —The coordinates of any pixel of the crack in the image; —The coordinates of another pixel of the crack in the image; Step 5, the crack length and the composite apparent density of the filler are used to correct the specific surface area; ; Where: —Corrected value of filler specific surface area, ; —Specific surface area of filler, ; —Average sieve hole classification residual percentage of filler, ; — synthetic wool volume density of the filler, ; Step six, calculate the cement absorption coefficient of the filler; ; Where: — filler absorption coefficient of cement; —Effective relative density of filler; —composite apparent density of the filler, ; — relative density of cement; Step seven, calculating the roadbed filler particle size; ; Where: —Grain size of roadbed filler, ; — filler absorption coefficient of cement; After D is calculated, take the difference between D and the sieve hole sizes of 0.075, 0.15, 0.3, 0.6, 1.18, 2.36, 4.75, 9.5, 13.2, and 16. The sieve hole size corresponding to the smallest difference is the final roadbed filler particle size.
2. According to the method for determining the particle size of roadbed filler by image processing as described in claim 1, it is characterized in that: Step 2 is as follows: (1) Match the structural element with the pixels in the image. If the structural element contains the image pixel, the image pixel value and the surrounding pixel values are enhanced. If the structural element does not contain the pixel, no processing is performed; (2) After the operation in (1), the result is subjected to Gaussian filtering; (3) Divide the shadow part by threshold segmentation, perform the maximum entropy threshold segmentation operation on the processing result of (2), and obtain the decision value of whether to segment, which is used as the segmentation threshold; (4) Calculate the standard deviation of pixel brightness in the non-shadow area and the average pixel brightness , the road image is divided into N = 256 regions of different brightness, Make the area Contains brightness value All pixels of , ; (5) Calculate the image pixel compensation coefficient; ; Where: —Image pixel compensation coefficient; —The average pixel brightness value of the non-shadow part; —The average pixel brightness value of the shadow part; —The standard deviation of pixel brightness in the non-shadow part; —The standard deviation of pixel brightness in the shadow part; (6) Compensate the brightness value of the pixels in the image; ; Where: —The brightness value of each pixel of the processed road surface image; —The brightness value of each pixel of the processed road surface image after compensation; (7) Binarize the image.
3. According to the method for determining the particle size of roadbed filler by image processing as described in claim 2, it is characterized in that: Step three is as follows: initialize the binarized image, process each pixel one by one in a predetermined order, and decide whether to perform stripping based on the situation of the surrounding pixels. If at least one pixel around a pixel is not an edge pixel, the pixel cannot be stripped, otherwise the pixel can be stripped. In the segmented binary image, the grayscale value of the retained points is assigned to 0, and the grayscale values of the remaining points are assigned to 255.
4. The method for determining the particle size of roadbed filler by image processing according to claim 3, characterized in that: Step 4: After the pixel scanning, if the grayscale value of a certain pixel is found to change from 255 to 0, then this pixel is considered to be the uppermost pixel of the crack; if the grayscale value of a pixel is found to change from 0 to 255, it is considered to be the lowermost pixel.
