CN1995968B - Method for extracting efficient gap characteristic parameter of drainage asphalt concrete - Google Patents

Abstract

The invention relates to a sewage asphalt concrete effective gap feature parameter extraction method from digital image of the cross section of the breakage. Stuffing the inside of the sewage asphalt concrete sample with white paste and cut the hardened testing sample. It uses the digital image analyzing software to intensify and divide the image or eliminates the noise, extracting features, getting effective gaps quantity N, each effective gap area Ai and outside perimeter. It then calculates the area ratio, average outside perimeter, average roundness and effective gap area frequency distribution of the effective gap area. The distribution feature parameter quantitative description of the effective gap can work as an important index for the design and also for the impact rules of the effective distribution gap.

Description

A kind of method for extracting efficient gap characteristic parameter of drainage asphalt concrete

Technical field

The present invention relates to the assay method of a kind of drainage asphalt concrete space feature, particularly a kind of method of from the digital picture of drainage asphalt concrete section, extracting its active porosity distribution characteristics parameter.

Background technology

The drainage asphalt concrete surface layer rainy day can be got rid of surface area water rapidly, reduces to dabble and spray, and has the surface of good cling property, can greatly strengthen traveling security.Simultaneously, it can also obviously reduce the noise on tire and road surface, improves the pollution of traffic to environment.The outstanding feature of drainage asphalt concrete be voidage up to about 20%, wherein comprise connectivity gap, semi-connected space and dead air space.Dead air space is independent closed space, can't absorb sound and draining.Therefore, active porosity specially refers to connectivity gap and semi-connected space, and the distribution characteristics of effective porosity and active porosity is the key that influences drainage asphalt concrete draining and anti-acoustic capability.At present, in the design mainly be the bulk volume relative density γ that measures the moulding marshal piece with volumetric method _f, then according to its theoretical maximum relative density γ of cooperation composition material density calculation when of asphalt _t, by formula $V_V=(1-\frac{{\mathrm{\γ}}_f}{{\mathrm{\γ}}_t})\×100\%$ Calculate the voidage V of drainage asphalt concrete _VThis method can only provide the volumn concentration in full space, can't truly reflect the distribution characteristics of inside concrete active porosity.In addition, in the existing drainage asphalt concrete mixing proportion design method, can not reflect the influence of the variation of mineral aggregate gradation and asphalt content, not have the relevant art means to measure this key index of distribution characteristics of active porosity to inside concrete space distribution characteristics.

Summary of the invention

At the problem that is difficult to measure drainage asphalt concrete active porosity distribution characteristics parameter, the present invention carries out special processing to the drainage asphalt concrete sample, utilize computer digital image to handle means, a kind of method of extracting the inner active porosity distribution characteristics of drainage asphalt concrete parameter is provided.

The present invention realizes according to the following steps:

(1) active porosity with drainage asphalt concrete sample inside pours into plain boiled water mud, and the specimen that thickness is 1～2cm is made in the back section of waiting to harden, so that the active porosity of inside concrete and skeleton structure present a contrast on gray scale.

(2) obtain the digital picture of test sample surface, use " cosmos32 multifunctional universal image analysis software " it is carried out the processing of figure image intensifying, image segmentation, noise removing, then binary image is carried out feature extraction, obtain the quantity N of active porosity, the area A of each active porosity _iWith outer perimeter L _i

(3) area occupation ratio of calculating active porosity, average outside perimeter, average circularity, and the frequency distribution of active porosity area.

Wherein, the meaning of each parameter and computing method are as follows more than:

The quantity N of active porosity: promptly measure the active porosity number that is identified in the zone.

The area occupation ratio a of active porosity: promptly the summation of each active porosity area accounts for the percent of measuring the regional integration area, $a=A/S=\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{A}_i/S$ (A _i: the area of each active porosity, i=1～N; S: the area of measuring the zone).

The average outside perimeter of active porosity Be the mean value of each active porosity outer perimeter, $\stackrel{\‾}{L}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{L}_i$ (L _i: the outer perimeter of each active porosity, i=1～N)

The average circularity of active porosity

Be the mean value of each active porosity circularity, $\stackrel{\‾}{C}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{C}_i$ (C _i: the circularity of each active porosity, i=1～N).Circularity $C_i=(4\mathrm{\π}\×A_i)/L_i^2,$ When object was circle, this value was maximum (=1), this value will diminish (circle=1, oval＜1, irregularly shaped＜1, big more from 1 concavo-convex degree far away more) when object is elongated.

The frequency distribution of active porosity area: promptly each active porosity is in the distributed number situation of different area scope.

The described white cement of making plain boiled water mud, its whiteness is more than or equal to 80.

The water cement ratio of described plain boiled water mud is controlled between 0.48～0.60.

The present invention utilizes image recognition technology to obtain the characteristic parameter of its inner section active porosity by the drainage asphalt concrete sample is handled, and can be quantitatively described by the distribution characteristics of these parameters to active porosity.These parameters can be used as an important indicator of reference in the match ratio design, also be further to inquire into the influence rule of match ratio, and the relation of research microscopic voids structure and macroscopical pavement performance is laid a good foundation to drainage asphalt concrete inside active porosity distribution characteristics.

Description of drawings

Fig. 1 is based on the workflow diagram that drainage asphalt concrete active porosity distribution characteristics is measured in image recognition

Fig. 2 is the drainage asphalt concrete sample that is placed in the die trial

Fig. 3 is the state after the die trial of placement sample pours into plain boiled water mud

Fig. 4 is through maintenance, the drainage asphalt concrete test specimen after the grout sclerosis

Fig. 5 is that test specimen is through the latter made specimen of section

Fig. 6 is the digital picture (RGB pattern) of test sample surface

Fig. 7 is through the gray level image after the mode switch

Fig. 8 is the later effect of expansion that image carries out dynamic lens

Fig. 9 is that image strengthens the effect after filter is handled

Figure 10 is that image carries out the effect after level and smooth filter is handled

Figure 11 is the bianry image after handling through image segmentation

Figure 12 carries out effect after noise removing is handled to image

Figure 13 demarcates decipherment distance

Figure 14 is the histogram of active porosity area

Embodiment

The invention will be further described below in conjunction with Fig. 1～Figure 13.

The drainage asphalt concrete sample can be made according to the related request of " bitumen mixture specimen method for making " in " highway engineering pitch and asphalt testing regulations (JTJ 052-2000) ", is of a size of φ 101.6mm * 63.5mm; Or from the drainage asphalt pavement core boring sampling, core diameter is 100mm, highly is 30mm～80mm.

Present embodiment mainly comprises following three steps:

1. the preparation of specimen and processing

Present embodiment is of a size of the drainage asphalt concrete right cylinder test specimen of φ 101.6mm * 63.5mm in prepared in laboratory, test specimen is placed in the die trial of 15cm * 15cm * 15cm, and shop, die trial bottom one deck rubble, as shown in Figure 2.

Adopting whiteness is 88 white cement, and making water cement ratio with the suitable quantity of water mix is 0.55 plain boiled water mud, and it is poured in the die trial, and liquid level exceeds the about 3cm of test specimen end face.Then, die trial is placed on about 300s that shakes on the shaking table, makes grout fully enter into the connectivity gap and the semi-connected space of test specimen inside, closed dead air space then can not enter plain boiled water mud, as shown in Figure 3.

Die trial is moved in the standard curing room, maintenance form removal after seven days, as shown in Figure 4.

Test specimen after the sclerosis is cut into the specimen that thickness is 1cm, as shown in Figure 5.

2. digital picture obtains and Flame Image Process

Adopt Sony DSC-H ₁The RGB mode digital image that the type digital camera is taken pictures and obtained test sample surface, as shown in Figure 6.Present embodiment adopts " Cosmos32 multifunctional universal image analysis software " that image is handled.

At first the digital picture with the RGB pattern is converted to grayscale mode, throws away all colours information of image, only comes the level of analog color by 256 grades of gray scales.Then, the drag and drop by mouse are provided with measured zone, in the present embodiment circular arc interface in to connect in the square frame of broken lines be selection area, as shown in Figure 7.

Selection area is carried out the processing of figure image intensifying, the expansion by dynamic lens, strengthen filter, level and smooth filter is realized.Earlier image is carried out the expansion of dynamic lens, parameter is that treatment effect as shown in Figure 8 automatically.Then, strengthen the processing of filter, adopt the form of emphasizing 8 directions, be of a size of 5 * 5, divisor is 1, and treatment effect as shown in Figure 9.Carry out the processing of level and smooth filter again, take simple smoothing, be of a size of 5 * 5, divisor is 25, and treatment effect as shown in figure 10.The purpose of figure image intensifying is to change the deep or light information that image had, and outstanding contrast is improved picture quality, makes clear picture, and feature is obvious, more helps Computer Identification and Analysis.

Selection area is carried out image segmentation, promptly measuring object thing (space) is carried out and extracted.The method that image segmentation adopts binaryzation to cut apart makes image binaryzation by assign thresholds, and threshold value is made as 175.The order read each gray values of pixel points and with threshold ratio, all gray-scale values are converted into black smaller or equal to 175 picture element, belong to object (space); All gray-scale values are converted into white greater than 175 picture element, belong to background (skeleton structure).Its mathematical description is

T ₀Be threshold value, u (x ₁, x ₂) be the gray-scale value of original image; G (x ₁, x ₂) be the gray-scale value of output image after the binary conversion treatment, as shown in figure 11.

Image is carried out the processing of noise removing, and object is a black, and 8 face the territory links, and the picture element number is made as 50, promptly removes the particle of size below 50 pixel counts in the image, as shown in figure 12.

Next adjust the distance and demarcate.Pin left mouse button and be dragged to certain distance, in the physical length that " actual range " lining input mouse drags, unit is " μ m ", then realizes the demarcation of adjusting the distance by the ratio of scheming to go up pixel and actual range, as shown in figure 13.

Carry out sequential scanning to measuring the zone, the statistics gray-scale value is the number of the object (space) formed of 255 picture element, promptly obtains the quantity of active porosity; The pixel count summation of adding up each inside, measuring object thing edge (comprising on the edge) promptly obtains the area of each active porosity; Add up the length sum of 8 chain codes of each object (space) boundary pixel, promptly obtain the outer perimeter of each active porosity.

3. calculate the distribution characteristics parameter of active porosity

By the area and the outer perimeter of each active porosity, calculate area occupation ratio, average outside perimeter, the average circularity of measuring active porosity in the zone, and statistics obtains the frequency distribution of active porosity area.The frequency distribution of the quantity by active porosity, area occupation ratio, average outside perimeter, average circularity and active porosity area realizes the quantitative description to the distribution characteristics of active porosity.

The mensuration area of present embodiment is 50.1283cm ², measurement result is as follows:

Characteristic parameter	Numerical value
		The quantity of active porosity (individual)	92
The area occupation ratio of active porosity (%)	10.58
		The average outside perimeter of active porosity (cm)	1.3407
The average circularity of active porosity	0.4504

The frequency distribution of active porosity area as shown in figure 14.

The quantity of active porosity and area occupation ratio have reflected that drainage asphalt concrete inside comprises the structure situation of the active porosity and the skeleton in connectivity gap, semi-connected space, for guaranteeing the functional of drainage asphalt concrete surface layer draining and noise reduction, satisfy the requirement of pavement performance simultaneously, the area occupation ratio of active porosity should be between 8～14%.And the average outside perimeter of active porosity, average circular kilsyth basalt have been levied the shape facility of active porosity, under the roughly the same situation of area occupation ratio, the average circularity of active porosity is more near 1, average outside perimeter is more little, it is the approaching more circle of shape of active porosity, help bringing into play the function of concrete draining and noise reduction more, the average circularity should be more than or equal to 0.45.The frequency distribution of active porosity area has reflected the formation situation of each active porosity area size, the active porosity ratio height that area is big, and then the performance of draining and noise reduction is better.

Claims (3)

1. method for extracting efficient gap characteristic parameter of drainage asphalt concrete, it is characterized in that: this method is carried out according to the following steps:

1) active porosity with drainage asphalt concrete sample inside pours into plain boiled water mud, and the specimen that thickness is 1～2cm is made in the back section of waiting to harden;

2) obtain the digital picture of test sample surface, using " cosmos32 multifunctional universal image analysis software " handles this image, at first optional one measure region S, image in the selection area is carried out the processing of figure image intensifying, image segmentation, noise removing, image after handling is carried out feature extraction, obtain the quantity N of active porosity, the area A of each active porosity _iWith outer perimeter L _i

3) frequency distribution of the area occupation ratio of calculating active porosity, average outside perimeter, average circularity and active porosity area;

Wherein, the meaning of each parameter and computing method are as follows more than:

The quantity N of active porosity: promptly measure the active porosity number that is identified in the zone;

The area occupation ratio a of active porosity: promptly the summation of each active porosity area accounts for the percent of measuring the regional integration area;

Wherein: A _iBe the area of each active porosity, i=1～N, S is for measuring the area in zone;

The average outside perimeter of active porosity Be the mean value of each active porosity outer perimeter, Wherein: L _iBe the outer perimeter of each active porosity, i=1～N;

The average circularity of active porosity

Be the mean value of each active porosity circularity,

Wherein: C _iBe the circularity of each active porosity, circularity

The frequency distribution of active porosity area: promptly each active porosity is in the distributed number situation of different area scope.

2. a kind of method for extracting efficient gap characteristic parameter of drainage asphalt concrete according to claim 1 is characterized in that: the whiteness of white cement of making plain boiled water mud is more than or equal to 80.

3. a kind of method for extracting efficient gap characteristic parameter of drainage asphalt concrete according to claim 1 is characterized in that: the water cement ratio of described plain boiled water mud is 0.48～0.60.

Images