CN103940381B - A kind of soil surface roughness measuring method - Google Patents

Abstract

The invention discloses a kind of soil surface roughness measuring method, shoot by soil surface roughness is measured plate, then the picture of shooting is processed, obtain the actual height value of each probe and calculate root-mean-square height and the correlation length characterizing soil surface roughness.The present invention obtains, by shooting, the difference in height that soil surface rises and falls, and it is convenient to gather, the shortest, measures efficiency high.

Description

A kind of soil surface roughness measuring method

Technical field

The invention belongs to remote sensing soil parameters field of measuring technique, more specifically, relate to a kind of soil thick Rugosity measuring method.

Background technology

Soil surface roughness is to characterize Soil Hydrological characteristic and affect an Important Parameters of soil property.? In microwave remote sensing, soil surface roughness is the key factor affecting soil Microwave scattering characteristics, is that soil is micro- Important Parameters in scattering of wave model, is also the main parameters of soil parameters inverting, it is therefore desirable to accurate evaluation With measurement soil surface roughness.Soil surface roughness is represented by root-mean-square height and correlation length, and these are two years old Individual parameter is both vertically and horizontally characterizing the fluctuating situation of soil surface, i.e. roughness respectively.

The measuring method of soil surface roughness generally has contact and contactless two ways.Contactless Method uses laser to measure, and the measurement apparatus cost used by this measuring method is of a relatively high.Contact Measuring method is that when measuring, instrument contacts with soil, and contact method mainly has chaining pin method, chain method and bar chi Method.

At present, soil surface roughness test is contact type measurement mode based on bar chi method mostly, although its Measurement apparatus cost used is relatively low, but needs to insert the device into inside soil when it is measured, and this is to soil surface Having certain destructiveness, process of therefore measuring takes time and effort, and the accuracy and speed measured all is difficult to meet Requirement.

Summary of the invention

It is an object of the invention to overcome the deficiencies in the prior art, it is provided that one is applicable to field soil roughness Measure, convenient data acquisition, measure the soil surface roughness measuring method that efficiency is higher, the shortest.

For achieving the above object, the soil surface roughness in the present invention measures plate, including being disposed vertically Base plate, some probes；

Described base plate is rectangle, a length of 0.8～1.2 meter, height be 0.4～0.6 meter, and at four Angle Position It is respectively equipped with an obvious cross mark；

Described some probe Vertical Uniforms are placed in before base plate, are cased with rounded cap, and the end in each probe bottom End contacts with soil surface, and can move up and down according to soil surface fluctuating；Tips of probes needlepoint form, visits The epimere of pin is obvious with base plate color contrast, and probe upper portion obvious with base plate color contrast The difference in height that length rises and falls more than soil surface；

Measured with the angle shot soil surface roughness looked squarely as much as possible by high-resolution digital camera Plate, it is thus achieved that the photo of reflection soil surface roughness, then obtains the pixel coordinate of cross mark, and root According to the vertical dimension that cross mark is actual, it is thus achieved that the vertical scale factors of pixel to distance；Thick acquisition is shone again The pixel vertical coordinate of each tips of probes make height value in sheet, obtains actual height according to vertical scale factors Angle value, this height value reflects the fluctuating situation of each probe position；Reality finally according to each probe Border height value calculates root-mean-square height and the correlation length characterizing soil surface roughness.

The measuring method of a kind of soil surface roughness, it is characterised in that comprise the following steps:

(1), soil surface roughness is measured plate smooth vertical and be placed on tested soil surface；

(2), at distance soil surface roughness measurement plate 4m-5m by high-resolution digital camera as far as possible With the angle shot roughness concentration plate looked squarely, it is thus achieved that the picture of a reflection soil surface roughness；

(3), by soil surface roughness measure plate smooth vertical and be placed into the diverse location of tested soil surface, Placement direction does not limits, but requires not exist between position overlapping region, and the method same by step (2) shoots, So repeat, it is thus achieved that 20 pictures more than or equal to reflection soil surface roughness；

(4), to each shoot thick soil surface roughness and measure the picture of plate, in the following ways Process:

The drawing software that 4.1), carries with any a image processing software, such as Windows on computers, ACDSee or Adobe Photoshop, will shoot dividing of the picture of thick soil surface roughness measurement plate Resolution is modified to suitable size, and amended picture pixels is preferably 1280*768, the most successively record soil table Surface roughness measures the pixel coordinate record successively of cross mark on four angles of plate；Then image procossing is utilized " cutting " instrument of software intercepts the picture comprising the complete upper portion of all probes；

4.2) image processing program, is utilized first to carry out intercepting the picture comprising the complete upper portion of all probes Gray scale and luminance filtering, and picture is carried out binary conversion treatment；Then the same value phenomenon being likely to occur is processed； Wherein, referring to that different probe top connects together with value becomes a point, or different value is that same tips of probes divides Two points are become；

4.3) vertical coordinate on each top in all probes complete upper portion picture is comprised, after extraction process, And guarantee that each tips of probes extracts a vertical coordinate, it is sequentially stored into the array of probe height；

4.4), according to the actual range between cross identifies on soil surface roughness measurement four angles of plate, calculate Vertical scale factors, completes the pixel conversion to actual range；

$Y_{\mathrm{scale}}=(\frac{L_{0\mathrm{lift}}}{L_{\mathrm{dlift}}}+\frac{L_{0\mathrm{right}}}{L_{\mathrm{dright}}})/2---\left(1\right)$

In formula, Y_scaleIt is vertical scale factors, L_0liftTwo, plate left side cross is measured for soil surface roughness Vertical dimension actual between knowledge, L_0rightFor real between on the right of soil surface roughness measurement plate, two crosses identify The vertical dimension on border；L_dliftShoot 2, left side cross mark in the picture of thick soil surface roughness measurement plate Between number of pixels, by step 4.1) the pixel vertical coordinate of 2 the cross marks in the left side that obtains subtracts each other Obtain, L_drightShoot the picture between the crosses mark of 2, the right in the picture of thick soil surface roughness measurement plate Element number；By step 4.1) the pixel vertical coordinate of 2 the cross marks in the limit, the right that obtains subtracts each other acquisition；

By step 4.3) each in the array that obtains represent that the vertical coordinate of probe height is multiplied by vertical scale factors, Obtain the actual height of each probe；

(5), by 20 each group probe height data obtained more than or equal to reflection soil surface roughness picture Sequence merges into a probe height tables of data, then according to probe height tables of data camber data meter successively Calculate root-mean-square height and the correlation length characterizing tested soil surface roughness:

Root-mean-square height computing formula is:

$s={\left\{\frac{1}{N-1}\right[\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{\left(z_i\right)}^2-N{\left(\stackrel{\‾}{z}\right)}^2\left]\right\}}^{1/2}---\left(2\right)$

$\stackrel{\‾}{z}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{z}_i$

In formula (2), s is root-mean-square height,For probe height tables of data camber data mean value, i represents The sequence number of altitude information, z_iRepresent i-th altitude information, probe height tables of data camber data count；

Correlation length computing formula is:

$\hat{V}\left(x_k\right)=\frac{1}{2(N-k)}\underset{i=1}{\overset{i=N-k}{\mathrm{\Σ}}}{(z_{i+k}-z_i)}^2$

$\widehat{\mathrm{\γ}}\left(x_k\right)=s^2-\hat{V}\left(x_k\right)$

$\widehat{\mathrm{\ρ}}\left(x_k\right)=\frac{\widehat{\mathrm{\γ}}\left(x_k\right)}{\widehat{\mathrm{\γ}}\left(0\right)}=\frac{1}{e}$

In formula,For semi-variety figure (semi-variogram), subscript k is the probe number at interval, x_kFor height Degrees of data z_i+kTo z_iBetween distance and equal to kl_p, l_pFor the spacing between adjacent probe；For covariance function,For auto-correlation function, l is correlation length, corresponding when auto-correlation function curve drops to 1/e for the first time Distance value x_kIt it is i.e. correlation length l of tested soil surface.

The object of the present invention is achieved like this:

If soil surface roughness of the present invention measures plate by being uniformly arranged at the base plate front vertical being disposed vertically Dry probe, each probe bottom is cased with rounded cap, and bottom contacts with soil surface, and can be according to soil surface Fluctuating moves up and down, and probe epimere is obvious with base plate color contrast.So by soil surface roughness Measure plate to shoot, then the picture of shooting is processed, obtain the actual height value of each probe also Calculate root-mean-square height and the correlation length characterizing soil surface roughness.Soil surface roughness measures plate Relatively simple for structure, size is little, it is simple to carry, it is adaptable to field soil roughness concentration；It addition, it is logical Crossing shooting and obtain the difference in height that soil surface rises and falls, it is convenient to gather, the shortest, measures efficiency high；Finally, Owing to there is rounded cap probe bottom, will not insert inside soil, it is to avoid destruction to soil surface.

Accompanying drawing explanation

Fig. 1 is that soil surface roughness of the present invention measures plate one detailed description of the invention structure chart；

Fig. 2 is soil surface roughness measuring method one detailed description of the invention flow chart of the present invention；

Fig. 3 is the root-mean-square height figure that one meter of 45 measurement data measuring length is extracted；

Fig. 4 is the correlation length figure that different sampling length extracts；

Fig. 5 is the auto-correlation function comparison diagram that soil is obeyed.

Detailed description of the invention

Below in conjunction with the accompanying drawings the detailed description of the invention of the present invention is described, in order to those skilled in the art is more Understand the present invention well.Requiring particular attention is that, in the following description, when known function and setting Perhaps, when the detailed description of meter can desalinate the main contents of the present invention, these are described in and will be left in the basket here.

Fig. 1 is that soil surface roughness of the present invention measures plate one detailed description of the invention structure chart.

In the present embodiment, as shown in Figure 4, soil surface roughness of the present invention measures including of plate: vertical The base plate 1 of placement, some probes 2；

Described base plate 1 is rectangle, and its color is black, a length of 1m, a height of 0.525m；At base plate 1 four Individual Angle Position is respectively equipped with an obvious cross mark 5,6,7,8, cross mark 5,6,7,8 conducts Coordinate reference points calculates for vertical scale factors, pixel unit is scaled physical length unit, specifically changes Calculate as shown in Equation (1).

Described some probes 2 Vertical Uniform is placed in before base plate 1, and probe 2 is a length of 400mm, directly Footpath is the rustless steel cylinder of 2.5mm-3mm, and base plate 1 above 110 probes 2, spacing is 6mm, so Can obtain abundant soil surface sampled point, probe 2 has sufficiently large measurement scope simultaneously.

Being cased with rounded cap 3 in each probe 2 bottom, and bottom contacts with soil surface, rounded cap 3 is 2.5mm-3mm Circular STAINLESS STEEL NUTS, the purpose adding STAINLESS STEEL NUTS is to make probe 2 can contact with soil surface, but not Can insert in soil.Probe 2 can rise and fall according to soil surface and move up and down, and the scope that moves up and down is for ± 10cm. Probe 2 top needlepoint form, the epimere of probe 2 is cased with the white hot draw 4, such base plate 1 black and probe The most clearly, the white hot draw 4 is more than the difference in height of soil surface fluctuating for the contrast of the white epimere of 2. In the present embodiment, a length of 8cm-10cm of the white hot draw 4, a diameter of 3.7mm, soil surface is thick With the white hot draw 4 as target when the difference in height that the i.e. soil surface of rugosity information rises and falls is extracted, black floor 1 Contrast substantially with white hot sleeve pipe 4, beneficially soil surface roughness information retrieval.

The support bar 9 at base plate 1 back side is welded in the back side of black floor 1, is used for supporting and fix base plate 1, makes It steadily stands on tested soil surface.

In the present embodiment, as it is shown in figure 5, soil surface roughness measuring method of the present invention, including as follows Step:

Step ST1: described soil roughness measurement plate is stably placed at tested soil surface, and requires soil Earth roughness concentration plate is perpendicular to tested soil surface, the support bar fixing soil then installed below with base plate Roughness concentration plate so that it is firm stable；

Step ST2: after soil roughness measurement plate is steadily determined, using high-resolution number at its 4m-5m Code-phase machine measures plate with the angle shot soil roughness looked squarely as far as possible, it is thus achieved that a reflection soil surface is thick The picture of rugosity；

Step ST3: soil surface roughness is measured plate and is placed into the diverse location of tested soil surface, put Put direction not limit, but require not exist between position overlapping region, with the side of same method, i.e. step 2 Method shoots, and so repeats, it is thus achieved that 20 pictures more than or equal to reflection soil surface roughness；

Step ST4: on computers with any a image processing software, the picture carried such as Windows is soft Part, ACDSee or Adobe Photoshop etc., will shoot the picture of soil surface roughness measurement plate Resolution is modified to suitable size, and amended picture pixels is preferably 1280*768；

Step ST5: soil surface roughness is measured the vertical seat of cross mark 5,6,7,8 on four angles of plate Mark recorded in the Excel file of named " Coordinate " successively, by the soil of row successively records photographing The coordinate of four cross marks of surface finish measurement plate picture, the corresponding soil surface of every a line is coarse Degree measures the coordinate of four cross marks on plate picture, and record form is as shown in table 1:

Table 1

In table 1, on four angles of every a line correspondence roughness concentration plate, cross mark is surveyed at soil surface roughness The vertical coordinate of template picture, unit: pixel number.

Step ST6: " cutting " instrument intercepting of utilization comprises instrument intercepting and comprises all probes complete epimere portion Point picture, the i.e. complete picture retaining all tips of probes, and with " 1,2,3 ..., n " literary composition Part title saves as extending the picture file of entitled " .jpg " successively.

Step ST7: utilize image processing program to comprise the complete upper portion of all probes to above-mentioned to intercepting First picture carries out gray scale and luminance filtering, and picture is carried out binary conversion treatment.

Step ST8: comprise the vertical of each top in all probes complete upper portion picture after extraction process Coordinate, and process the same value or different value phenomenon being likely to occur, it is ensured that each needle point extracts a vertical coordinate, To the array of a series of expression probe height, the corresponding number of picture that every soil surface roughness measures plate Group, save as named " 1,2,3 ..., n " txt notepad file, these arrays are merged into One tables of data, as the basis of calculating soil surface roughness, is saved in " All value.txt " file；

Step ST9: the Excel file " Coordinate " that reading step ST5 preserves, thick according to soil surface Rugosity measures the actual range on four angles of plate between cross identification point, calculates vertical scale factors.Will The all arrays preserved in step 8 are multiplied by the proportionality coefficient of each calculating, complete pixel to actual range Conversion, conversion method is as shown in Equation (1).

In the present embodiment, L_0lift=289.2mm, L_0right=289.0mm is that soil surface roughness measures the left side on plate And the vertical dimension between each two the cross identification points in the right, for given value, L_dliftAnd L_drightThe picture left side and the right side Number of pixels between each 2 the cross identification points in limit, is identified by left and right, 2, both sides, the right cross Pixel vertical coordinate subtracts each other acquisition.

Step ST10: for avoiding the different impact on measurement result of probe length, need during measurement soil table Surface roughness is measured plate and is calibrated, and during calibration, soil surface roughness is measured plate and is vertically disposed in level On smooth surface, extract the coordinate of tips of probes each point according to the method described above, save as an account herein Part " Calibration value.txt ", the tips of probes coordinate literary composition that then will obtain when measuring soil surface roughness The array of part " 1.txt, 2.txt, 3.txt ... " is individually subtracted this calibration file " Calibration value.txt " Corresponding array, completes calibration, i.e. eliminates the systematic error in measuring.

Step ST11: high more than or equal to each group probe that reflection soil surface roughness picture obtains by 20 Degrees of data sorts successively and merges into a probe height tables of data, then according to probe height tables of data camber Data calculate root-mean-square height and the correlation length characterizing tested soil surface roughness.

Data processed result is analyzed

At present when measuring and calculate soil roughness, do not have clear and definite for sampling length and measurement point interval Define, and the two parameter directly affects the result of calculated soil roughness, the measurement side of this patent Method solves this problem well.Experimental selection measures 45 meters in the same direction in same plot, as Calculate root-mean-square height and the data source of correlation length under different sampling length.

Fig. 3 gives the extraction result of root-mean-square height, and a point in figure represents 1 survey measuring position The root-mean-square height that amount data are extracted, each position measurement length is 1 meter, with the soil table of the present embodiment Surface roughness is measured plate and is gathered the fluctuating situation of soil.45 points in Fig. 1 represent the mean square of 45 positions Root height, as can be known from Fig. 1, if choosing the measurement length of 1 meter, the value of obtained root-mean-square height It is the most unstable, it is difficult to determine the root-mean-square height of tested soil surface, if the multiple positions gathered are carried out Calculate root-mean-square height after combination again, be equivalent to increase sampling length, can make by altitude information extract equal Root height tends towards stability, and however it is necessary that the minima determining sampling length and the relation extracting result, in order to Determine pendulous frequency, i.e. sampling length in the measurements.The data now measuring position acquisition by 45 are combined, I.e. increase sampling length, extract the root-mean-square height of tested soil surface the most again, then can improve extraction knot The stability of fruit, and determine sampling length.

Fig. 4 gives the root-mean-square height variation relation with sampling length, works as sampling length as can be known from Fig. 2 Time shorter, obtained root-mean-square height is unstable, it is impossible to effectively characterize the root-mean-square of tested soil surface Highly, after sampling length is more than 20 meters, root-mean-square height basically reaches stationary value.

In like manner, Fig. 5 gives the correlation length change curve with sampling length, according to the change of curve in Fig. 5 Change trend, after sampling length reaches 20 meters, correlation length tends towards stability substantially.If measurement can not connect Continuous measurement 20 meters, i.e. can not be continuously along the picture of straight line 20 roughness concentration plates of shooting, can be along arbitrarily Direction gathers several data of 1 meter, is then combined, and effect is identical.

Although detailed description of the invention illustrative to the present invention is described above, in order to this technology neck Artisans understand that the present invention, it should be apparent that the invention is not restricted to the scope of detailed description of the invention, to this From the point of view of the those of ordinary skill of technical field, as long as various change limits in appended claim and determines In the spirit and scope of the present invention, these changes are apparent from, and all utilize the invention of present inventive concept Create all at the row of protection.

Claims (2)

1. the measuring method of a soil surface roughness, it is characterised in that comprise the following steps:

(1), soil surface roughness is measured plate smooth vertical and be placed on tested soil surface；

Wherein, described soil surface roughness is measured plate and is included the base plate being disposed vertically, some probes；Institute Stating base plate is rectangle, a length of 0.8～1.2 meter, height be 0.4～0.6 meter, and set respectively at four Angle Position There is an obvious cross mark；Described some probe Vertical Uniforms are placed in before base plate, at each probe Bottom is cased with rounded cap, and bottom contacts with soil surface, and can move up and down according to soil surface fluctuating；Visit Pin top needlepoint form, the epimere of probe is obvious with base plate color contrast and bright with base plate color contrast The difference in height that the length of aobvious probe upper portion rises and falls more than soil surface；

(2), at distance soil surface roughness measurement plate 4m-5m by high-resolution digital camera as far as possible Plate is measured, it is thus achieved that the figure of a reflection soil surface roughness with the angle shot soil surface roughness looked squarely Sheet；

(3), by soil surface roughness measure plate smooth vertical and be placed into the diverse location of tested soil surface, Placement direction does not limits, but requires not exist between position overlapping region, and the method same by step (2) shoots, So repeat, it is thus achieved that more than or equal to the picture of 20 reflection soil surface roughness；

(4) soil surface roughness, to each shot measures the picture of plate, enters in the following ways Row processes:

4.1), with any a image processing software, the soil surface roughness shot is measured plate on computers The resolution of picture be modified to suitable size, amended picture pixels is 1280*768, record the most successively Soil surface roughness measures the pixel coordinate of cross mark on four angles of plate；Then utilize image procossing soft " cutting " instrument of part intercepts the picture comprising the complete upper portion of all probes；

4.2) image processing program, is utilized first to carry out intercepting the picture comprising the complete upper portion of all probes Gray scale and luminance filtering, and picture is carried out binary conversion treatment；Then the same value or different value being likely to occur is processed Phenomenon；Wherein, referring to that different probe top connects together with value becomes a point, or different value is same probe Top divide into two points；

4.3) vertical coordinate on each top in all probes complete upper portion picture is comprised, after extraction process, And guarantee that each tips of probes extracts a vertical coordinate, it is sequentially stored into the array of probe height；

4.4), according to the actual range between cross identifies on soil surface roughness measurement four angles of plate, calculate Vertical scale factors, completes the pixel conversion to actual range；

$Y_{scale}=(\frac{L_{0lift}}{L_{dlift}}+\frac{L_{0right}}{L_{dright}})/2---\left(1\right)$

In formula, Y_scaleIt is vertical scale factors, L_0liftTwo, plate left side cross is measured for soil surface roughness Vertical dimension actual between knowledge, L_0rightFor real between on the right of soil surface roughness measurement plate, two crosses identify The vertical dimension on border；L_dliftShoot 2, left side cross mark in the picture of thick soil surface roughness measurement plate Between number of pixels, by step 4.1) the pixel vertical coordinate of 2 the cross marks in the left side that obtains subtracts each other Obtain, L_drightShoot the picture between the crosses mark of 2, the right in the picture of thick soil surface roughness measurement plate Element number；By step 4.1) the pixel vertical coordinate of 2 the cross marks in the limit, the right that obtains subtracts each other acquisition；

By step 4.3) each in the array that obtains represent that the vertical coordinate of probe height is multiplied by vertical scale factors, Obtain the actual height of each probe；

(5), will be greater than each group probe height data obtained equal to 20 reflection soil surface roughness pictures Sequence merges into a probe height tables of data, then according to probe height tables of data camber data meter successively Calculate root-mean-square height and the correlation length characterizing tested soil surface roughness:

Root-mean-square height computing formula is:

$s={\{\frac{1}{N-1}\[\underset{i=1}{\overset{N}{\Σ}}{\left(z_i\right)}^2-N{\left(\stackrel{\‾}{z}\right)}^2\]\}}^{1/2}---\left(2\right)$

$\stackrel{\‾}{z}=\frac{1}{N}\underset{i=1}{\overset{N}{\Σ}}{z}_i$

In formula (2), s is root-mean-square height,For probe height tables of data camber data mean value, i represents The sequence number of altitude information, z_iRepresent i-th altitude information, probe height tables of data camber data count；

Correlation length computing formula is:

$\hat{V}\left(x_k\right)=\frac{1}{2(N-k)}\underset{i=1}{\overset{i=N-k}{\Σ}}{(z_{i+k}-z_i)}^2$

$\widehat{\mathrm{\γ}}\left(x_k\right)=s^2-\hat{V}\left(x_k\right)$

$\widehat{\mathrm{\ρ}}\left(x_k\right)=\frac{\widehat{\mathrm{\γ}}\left(x_k\right)}{\widehat{\mathrm{\γ}}\left(0\right)}=\frac{1}{e}$

In formula,For semi-variety figure, subscript k is the probe number at interval, x_kFor altitude information z_i+kTo z_iBetween Distance and equal to kl_p, l_pFor the spacing between adjacent probe；For covariance function,For auto-correlation function, L is correlation length, the distance value x corresponding when auto-correlation function curve drops to 1/e for the first time_kIt is i.e. tested Correlation length l of soil surface.

The measuring method of soil surface roughness the most according to claim 1, it is characterised in that also wrap Include scaling step, soil surface roughness is measured plate and is vertically disposed on the surface that level is smooth, according to upper The method of stating extracts the coordinate of tips of probes each point, by step (1) to step 4.3) method obtain one group Probe height array is as calibration probe height array, when measuring, step 4.3) probe height that obtains In array, the height of each probe also to deduct the probe height that calibration probe height array is corresponding.