Real-time evaluation method for gravel coverage rate of gravel seal construction site

Abstract

The invention relates to a gravel coverage rate real-time evaluation method for a gravel seal construction site. The method comprises the following steps: acquiring a to-be-detected lane gravel seal image; gray processing is carried out on the gravel seal coat image; highlight areas obtained in the global processing stage and the local processing stage are eliminated from the image; carrying out binarization processing on the image after highlight elimination; and obtaining the section gravel coverage rate. The method has the beneficial effects that based on the principle that the RGB value of the highlight area is the highest, the broken stone is the second time and the asphalt is the smallest, the reflection area in the image is effectively eliminated through the two-stage reflection removal algorithm, the influence of the highlight area on coverage rate detection in the detection process is reduced, and the detection precision of the broken stone coverage rate is improved; a method for determining a binarization threshold is provided, binarization processing is carried out on the detailed solution of the gravel seal map through a maximum between-class variance method, and then a gravel area and a non-gravel area are divided.

Description

Technical Field

[0001] The present invention belongs to the technical field of road construction, and particularly relates to a method for real-time evaluation of the gravel coverage rate at the construction site of a chip seal. Background Art

[0002] A chip seal is formed by spreading asphalt and gravel on a subbase through special equipment and then compacting to form a single-layer asphalt gravel seal. The surface gravel forms an interlayer interlocking effect with the upper layer structure to enhance the interlayer bonding effect. It has advantages such as long-lasting waterproofness, excellent skid resistance, and good economy. The interlayer gravel is bonded to the subbase through asphalt and bonded to the asphalt mixture layer by interlocking to increase the interlayer bonding between the subbase and the asphalt mixture layer. During the actual construction process, if the gravel spreading amount is too small, the interlocking effect is poor and the interlayer bonding strength cannot reach the optimal state; if the gravel spreading amount is too large, an interlayer composed of gravel and asphalt is formed between the layers and the interlocking effect cannot be formed. Therefore, in the specifications, the gravel coverage rate is evaluated by the area of gravel coverage per unit area, and the standard is 70% - 90%.

[0003] Patent numbers CN215758378U, CN216107928U, and CN216107927U introduce different types of chip seal construction equipment; patent numbers CN117468294A and CN115160769A introduce chip seals under different binders

[0004] In the prior art for evaluating the gravel coverage rate, CN217104630U introduces an indoor evaluation method for the gravel coverage rate, which is only suitable for indoor evaluation. CN115880255A and CN117779558A introduce two detection methods for the coverage rate during the construction process of chip seals, but they do not solve the problem of calculation errors in the coverage rate caused by high light due to sunlight reflection during on-site detection.

[0005] In the prior art, the evaluation of the gravel coverage rate at the chip seal construction site mostly uses manual observation or simple image processing methods, which have deficiencies such as strong subjectivity, low accuracy, and inability to detect in real time. Especially under light conditions, the asphalt material in the chip seal will form a reflection, resulting in an increase in the calculation error of the coverage rate.

[0006] Therefore, there is an urgent need for a method that can accurately and real-time evaluate the gravel coverage rate at the chip seal construction site. Summary of the Invention

[0007] The purpose of the present invention is to overcome the deficiencies in the prior art and provide a method for real-time evaluation of the gravel coverage rate at the chip seal construction site.

[0008] This method for real-time evaluation of the gravel coverage rate at the chip seal construction site includes the following steps:

[0009] Step 1: Collect the chip seal image of the lane to be inspected;

[0010] Step 2: Perform grayscale processing on the chip seal image, and perform median filtering and noise reduction processing on the grayscale image;

[0011] Step 3: Obtain the global highlight area through the global processing stage, and detect the part with weak highlight intensity in the image through the local processing stage; Eliminate the highlight areas obtained in the global processing stage and the local processing stage from the image;

[0012] Step 4: Calculate the segmentation threshold T by the Otsu method, and perform binary processing on the image after eliminating the highlights;

[0013] Step 5: Calculate the proportion of white pixel points after binary processing to obtain the chip coverage rate of the cross-section.

[0014] Preferably, in Step 3, in the global processing stage, calculate the ratios of the intensities of the red, green, and blue channels at the 98th percentile to the intensity value of the grayscale image at the 98th percentile respectively, to obtain the control values r RE 、r GE and r BE , when c R (x) at the pixel point x position > r RE ·T1, c G (x) > r GE ·T1, c B (x) > r BE ·T1 and c E (x) > T1, then x is the position of the highlight pixel point, c R (x), c G (x), c B (x) are the red, green, and blue intensity values of the pixel point at each x position respectively, and T1 is the color intensity threshold.

[0015] Preferably, in Step 3, in the local processing stage, first perform median filtering on the image, calculate the maximum value of the intensity ratios of the three channels at each pixel point x before and after filtering, and when the above maximum value is greater than the threshold then this area is the highlight area.

[0016] Preferably, in Step 3, after eliminating the highlight areas obtained in the global processing stage and the local processing stage from the image, perform smoothing processing using neighboring pixels to obtain the image after eliminating the highlights.

[0017] Preferably, in Step 4, first calculate the probability of each grayscale image pixel at each gray level, and then calculate the between-class variance σ 2(k), and find the between-class variance σ 2 (k) The largest k value is used as the segmentation threshold T and binarization is performed.

[0018] Preferably, in step five, the binary image of the gravel seal is equally divided into several blocks, the gravel coverage rate of each block is calculated respectively, and the uniformity of gravel spreading of the entire construction lane is analyzed in combination with the mileage of the gravel spreading vehicle.

[0019] Preferably, in step three, T1 is 245.

[0020] The beneficial effects of the present invention are:

[0021] 1) Based on the principle that the RGB value of the highlight area is the highest, followed by gravel and the smallest of asphalt, the present invention effectively eliminates the reflective area in the image through a two-stage de-reflection algorithm, reduces the influence of the highlight area on the coverage detection during the detection process, and improves the detection accuracy of the gravel coverage.

[0022] 2) The present invention proposes a method for determining a binarization threshold value, binarizes the detailed explanation of the chip seal map by the maximum inter-class variance method, and then separates the chip seal area and the non-chip seal area, and by dividing the calculation area unit of the chip seal coverage, the construction lane can be fully inspected in real time, and a vehicle equipped with a high-definition camera is used for image acquisition, combined with Matlab software for image processing and analysis, to achieve real-time and full-process detection of the chip seal construction site, and provide a reliable reference for the evaluation of the effect of chip spread. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 The crushed stone coverage rate calculation implementation process provided by the present invention;

[0024] Figure 2 An image of a chip seal image subjected to de-reflection processing by a two-stage algorithm provided by the present invention;

[0025] Figure 3 A comparison of black and white images of the chip seal before and after the reflective treatment provided by the present invention;

[0026] Figure 4 A three-equal-section image of the binary image of the chip seal provided by the present invention;

[0027] Figure 5 This is a graph showing the uniformity evaluation results of gravel spreading on a construction lane provided by the present invention. DETAILED DESCRIPTION

[0028] The present invention is further described below in conjunction with embodiments. The description of the following embodiments is only used to help understand the present invention. It should be noted that for ordinary persons in the art, without departing from the principle of the present invention, the present invention can also be modified in some ways, and these improvements and modifications also fall within the scope of protection of the claims of the present invention.

[0029] Embodiment 1

[0030] As an example, Figure 1 As shown, the real-time evaluation method of the crushed stone coverage rate at the crushed stone seal construction site includes the following steps:

[0031] Step 1: Collecting the gravel seal image of the lane to be inspected;

[0032] Step 2: grayscale the gravel seal layer image, and perform median filtering noise reduction on the grayscale image;

[0033] Step 3: Obtain the global highlight area through the global processing stage. In the global processing stage, first take the color intensity threshold T1, calculate the ratio of the red, green and blue channel intensity at the 98th percentile to the grayscale image intensity value at the 98th percentile, and obtain the control value r of the three channels. RE 、r GE and r BE , when the pixel point x position c R (x)>r RE ·T1, c G (x)>r GE ·T1, c B (x)>r BE ·T1 and c E (x)>T1, then x is the position of the highlight pixel, c R (x), c G (x), c B (x) are the red, green, and blue intensity values ​​of the pixel at each x position.

[0034] The part with weaker highlight intensity in the image is detected through the local processing stage; in the local processing stage, the image is first median filtered to calculate the maximum value of the intensity ratio of the three channels at each pixel point x before and after filtering. This area is the highlight area.

[0035] The highlight areas obtained in the global processing stage and the local processing stage are removed from the image.

[0036] Step 4: Calculate the segmentation threshold T by using the maximum inter-class variance method, and perform binarization on the image after highlight removal;

[0037] Step 5: Calculate the proportion of white pixels after binarization to obtain the gravel coverage rate of the cross section.

[0038] Embodiment 2

[0039] As another embodiment, this embodiment 2 proposes, based on the embodiment 1, a more specific method for real-time evaluation of the gravel coverage rate at the gravel seal construction site:

[0040] In step 1, the step of obtaining the image of the gravel seal layer of the lane to be inspected includes:

[0041] Equipment preparation: First, prepare a vehicle equipped with a high-definition camera to ensure that the camera can capture clear and stable images of the chip seal surface. At the same time, ensure that the vehicle's driving speed is controllable so that it can maintain a constant speed while collecting images.

[0042] Image acquisition: Start Matlab software and use the videoinput function to open and read the USB serial port of the camera. Then, set the getsnapshot function to acquire an image of the chip seal layer every 1 second. During the acquisition process, pay attention to controlling the speed of the vehicle and avoid braking or starting to reduce damage to the chip seal layer.

[0043] In step 2, the steps of grayscale and noise reduction processing on the chip seal layer image are as follows:

[0044] Grayscale processing: Use the rgb2gray function in Matlab software to grayscale the collected gravel seal layer image. The processing formula is Gray = 0.299R + 0.587G + 0.114B, where R, G, B are the red, green and blue channel values ​​of each pixel, respectively, and Gray is the grayscale value of each pixel.

[0045] Noise reduction: Perform median filtering on the grayscale image. Use the medfilt2 function to perform median filtering on the grayscale image to remove noise and spots in the image.

[0046] Step 3: De-glare the chip seal image

[0047] Global processing stage: Process global reflections (highlights). The global image contains asphalt, gravel and highlight areas. The highlight area has the highest RGB value, followed by gravel, and the asphalt has the lowest. In order to prevent the gravel from being eliminated during the elimination of the highlight area, the R, G, and B values ​​of each pixel are judged to improve the processing accuracy.

[0048] Since the intensity and angle of sunlight will affect the RGB value of the image in actual application, the global highlight area is first obtained by using the ratio of the red, green and blue channel intensity to the grayscale image intensity and the color intensity threshold T1. Through trial calculation, the control value r is calculated by using the red, green and blue channel intensity at the 98th percentile and the grayscale image intensity value at the 98th percentile. RE 、r GE and r BE The calculation formula is as follows:

[0049]

[0050] Then determine the highlight pixel position by the following formula, where x is the highlight pixel position

[0051] c R (x)>r RE ·T1 and c G (x)>r GE ·T1 and c B (x)>r BE ·T1 and c E (x)>T1

[0052] In this embodiment, based on experience, the T1 value is taken as 245.

[0053] Local processing stage: Each given pixel is compared with the color of a smooth non-specular surface at the pixel location, which is estimated by local image statistics. This stage aims to detect parts of the image with weak highlight intensity. The image is first median filtered to obtain the red, green, and blue intensity values ​​of the pixel at each x position. and Then, the red, green, and blue channel intensities c of the pixels at each x position in the original image are used. R (x), c G (x), c B (x) value, calculate the intensity ratio of the three channels before and after filtering, and take the maximum value.

[0054]

[0055] Then set a threshold when This indicates that the x position is a highlight area.

[0056] After two stages of highlight detection, the highlight area is eliminated and then smoothed using neighboring pixels to obtain an image with the highlights eliminated.

[0057] It should be noted that the parts in this embodiment that are the same or similar to those in the first embodiment can be referenced to each other and will not be described in detail in this application.

[0058] Example 3

[0059] As another example, this Example 3 is proposed on the basis of Examples 1 and 2, and a more specific real-time evaluation method for the gravel coverage rate at the gravel seal construction site is as follows:

[0060] In Step 4, the method for binarizing the gravel seal image is specifically as follows:

[0061] Calculate the gray level probability: First, calculate the probability p(q) of each pixel in the gray image being at each gray level according to the following formula, where n represents the total number of pixels, q represents the gray level of the image, and n q represents the total number of pixels at the q gray level.

[0062]

[0063] Determine the segmentation threshold: Determine a threshold k in advance, and divide the image into two groups of gray levels, namely m0 and m1. m0 corresponds to the pixels with gray levels [0, 1,..., k - 1], and m1 corresponds to the pixels with gray levels [k, k + 1,..., L - 1]. Use the maximum between-class variance method to determine the segmentation threshold T for binarization.

[0064] The between-class variance of the image gray level is expressed as:

[0065] σ 2 (k) = w0(μ0 - μ T ) 2 + w1(μ1 - μ T ) 2

[0066] In the formula:

[0067] Solve for the threshold k that maximizes the between-class variance, which is the required segmentation threshold T.

[0068] Binarization processing: Finally, binarize the image according to the segmentation threshold T to distinguish the gravel area and the non-gravel area in the image.

[0069] In Step 5, calculate the gravel coverage rate and analyze the spreading uniformity of the entire construction lane:

[0070] Calculate the gravel coverage rate: In the binarized image, white pixel points represent the gravel area, and black pixel points represent the non-gravel area. By calculating the number of white pixel points n w and the total number of pixels n in the picture, the cross-section gravel coverage rate can be calculated, and the gravel coverage rate = n w / n.

[0071] Analysis of spreading uniformity: In this embodiment, in order to analyze the uniformity of crushed stone spreading, the binary image of the crushed stone seal coat collected is equally divided. As Figure 4 shown, a binary image is equally divided into three parts, and the crushed stone coverage rate in different regions of the whole image can be analyzed to analyze the uniformity of crushed stone spreading.

[0072] As Figure 5 shown, combined with the driving mileage of the crushed stone spreader, calculate the crushed stone coverage rate of each block respectively. If the crushed stone coverage rates of each block are not much different and remain stable or slightly fluctuate with the increase of driving mileage, it can be considered that the crushed stone spreading is uniform. Otherwise, the spreading process needs to be adjusted to improve the spreading uniformity.

[0073] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. For the same and similar parts among the various embodiments, reference can be made to each other.

Claims

1. A real-time evaluation method for the gravel coverage rate at a gravel seal construction site, characterized in that: The following steps are involved: Step 1: Collecting the gravel seal image of the lane to be inspected; Step 2: grayscale the gravel seal layer image, and perform median filtering noise reduction on the grayscale image; Step 3: Obtain the global highlight area through the global processing stage, and detect the part with weaker highlight intensity in the image through the local processing stage; Eliminate the highlight areas obtained in the global processing stage and the local processing stage from the image; Step 4: Calculate the segmentation threshold T by using the maximum inter-class variance method, and perform binarization on the image after highlight removal; Step 5: Calculate the proportion of white pixels after binarization to obtain the gravel coverage rate of the cross section.

2. The real-time evaluation method for the gravel coverage rate at the gravel seal construction site according to claim 1 is characterized in that: In step 3, in the global processing stage, the ratios of the red, green and blue channel intensities at the 98th percentile to the grayscale image intensity value at the 98th percentile are calculated to obtain the control values ​​r of the three channels. RE 、r GE and r BE , when the pixel point x position c R (x)>r RE ·T1, c G (x)>r GE ·T1, c B (x)>r BE ·T1 and c E (x)>T1, then x is the position of the highlight pixel, c R (x), c G (x), c B (x) are the red, green and blue intensity values ​​of the pixel at each x position, and T1 is the color intensity threshold.

3. The real-time evaluation method for the gravel coverage rate at the gravel seal construction site according to claim 1, characterized in that: In step 3, in the local processing stage, the image is first median filtered to calculate the maximum value of the ratio of the three channel intensities at each pixel x before and after filtering. This area is the highlight area.

4. The real-time evaluation method for the gravel coverage rate at the gravel seal construction site according to claim 1, characterized in that: In step three, after the highlight areas obtained in the global processing stage and the local processing stage are removed from the image, the neighborhood pixels are used for smoothing to obtain an image after the highlights are removed.

5. The real-time evaluation method for the gravel coverage rate at the gravel seal construction site according to claim 1, characterized in that: In step 4, the probability of each grayscale image pixel at each grayscale level is first calculated, and then the inter-class variance σ of the image grayscale level is calculated 2 (k), and find the between-class variance σ 2 (k) The largest k value is used as the segmentation threshold T and binarization is performed.

6. The real-time evaluation method for the gravel coverage rate at the gravel seal construction site according to claim 1, characterized in that: In step 5, the binary image of the gravel seal is divided into several blocks, and the gravel coverage rate of each block is calculated. Combined with the mileage of the gravel spreader, the uniformity of gravel spreading in the entire construction lane is analyzed.

7. The real-time evaluation method for the gravel coverage rate at the gravel seal construction site according to claim 1, characterized in that: In step 3, T1 is set to 245.

Images