AU2020103570A4 - Grassland soil degradation evaluation method - Google Patents

Abstract

The present invention belongs to the technical field of soil detection, and discloses a grassland soil degradation evaluation method, which includes: selecting an evaluation target and completing automatic identification; inputting original data of target land; evaluating part of the original data of the target land; evaluating basic characteristics of the target land; evaluating homogeneity target land; and evaluating complexity target land. The present invention provides reference data for evaluating soil conditions suitable for a local agricultural environment and local agro-ecological characteristics. The present invention is suitable for instructing crop species, varieties and production technologies of the local agricultural ecosystem. The present invention provides a basis for regulating and controlling the ecological agriculture and intensive agriculture combining the local ecological system and agricultural economy. The present invention provides a basis for improving the erosion of the degraded agricultural soil, thereby guaranteeing the stabilization of the agricultural ecosystem and the stability of the agricultural environment. Drawings of Description S101 Select an evuaion target and complete automatic identification S102 Inpu t original data of target land 103 Evaluate part of the original data of the target land S104 Evaluate basic characteristics of the target land I - - - - - - ----------------------------S 1 0 5 Evaluate homogeneity target land S106 Evaluate complexity target land Fig. 1 1

Description

Drawings of Description

S101 Select an evuaion target and complete automatic identification

S102 Inpu t original data of target land

103 Evaluate part of the original data of the target land

S104 Evaluate basic characteristics of the target land I

- - - - - - ----------------------------S 1 0 5 Evaluate homogeneity target land

S106 Evaluate complexity target land

Fig. 1

Description

GRASSLAND SOIL DEGRADATION EVALUATION METHOD Technical Field

The present invention belongs to the technical field of soil detection, and particularly relates to a grassland soil degradation evaluation method.

Background

Natural conditions of grasslands in Xinjiang include climate, landform soil, hydrology, etc., which are the basic conditions for the formation of the grasslands. Xinjiang is located in the northwest of China. Its geographical position is on east longitude from 73°40' to 96°18' and northern latitude from 35°40' to 49°50', and it is adjoined to Gansu, Qinghai and Tibet on the east and south respectively. Xinjiang is neighbored to India, Pakistan, Afghanistan, Soviet Union and Mongolian People's Republic on the southwest and northeast. It is 1900 kilometers from east to west. It is 1500 kilometers from south to north and has a total area of 1.65 million square kilometers, accounting for one sixth of the total area in China. It has mountains, hills, plains, deserts, etc. Physiographically, Xinjiang can be basically divided into five major geomorphological units. There are Altay Mountains, belonging to the Altay fold-fracture mountain, and having a stepped landform on the southern slope rising in steps. Controlled by the westerly circulation, it is humid on the west and relatively dry on the east. Affected by the altitude, the precipitation is 500 mm in the alpine region and is 200-300 mm in the low-relief terrain. Therefore, Altay Mountain meadows and shrub grasslands are formed. Tianshan Moutains traverse the central part of Xinjiang and have complicated mountain structures. There are many fracture subsidence basins among the mountains, which are divided into the northern and southern slopes of Tianshan Mountains. The landform is complicated, and the land occupation area is large. Affected by westerly circulation, the northern slope is humid, the western region is relatively humid, and the eastern region is arid. With the change of sea-level elevation, meadows, pastures and desert grasslands are formed. Kunlun Mountains belong to the Kunlun fold zone and have large uplifts and towering ridges. The moisture from the Indian Ocean is blocked by the Himalayas and Kara Kunlun Mountains. The northern region is affected by the Taklimakan Desert, which constitutes the formation of mountainous barren and alpine desert grasslands. There are also some grasslands in alpine regions. The Junggar Basin and the Tarim Basin are two large Mesozoic and Cenozoic intermountain depressions. Due to the

Description

obstruction of peripheral mountains, the westerly circulation and the circulation moisture from the Indian Ocean are difficult to enter. However, the Junggar Basin is slightly better than the Tarim Basin. Generally speaking, it is still extremely arid, which constitutes the formation of plain desert grassland. Furthermore, there are also western mountains of Junggar. The mountains are low, are composed of fault-block mountains with stepped denuded surfaces, fault valleys and basins and have little rain, which constitutes the development of the pasture grassland. The Pamir Plateau is on the junction of the Kunlun Mountains and Tianshan Mountains. There are flat plateau areas that constitute the formation of alpine desert grasslands. From the perspective of climatic conditions, Xinjiang is located in the center of Eurasia and surrounded by high mountains and has relatively closed terrain. In terms of climate classification, it belongs to a temperate continental arid climate. Due to the vast territory and complex terrain, Xinjiang is rich and varied in thermal conditions from south to north and from basin to high mountains. There is not only a warm temperate zone, but also a temperate climate on the north of the Tianshan Mountains. Due to the influence of climate, a temperate desert-mountain pasture is formed in northern Xinjiang In southern Xinjiang, warm-temperate desert-mountain pastures, Pamirs plateau and Kunlun alpine desert pastures are formed. Although vast area of Xinjiang is arid and semi-arid and has the annual average precipitation about 150 mm, an overall distribution trend is as follows: Northern Xinjiang has more precipitation than Southern Xinjiang, the west has more precipitation than the east, the mountainous area has more precipitation than basins, the mountainous windward slopes have more precipitation than the leeward slopes, and the periphery of the basins has more precipitation than the center. The annual precipitation in plains and basins is relatively small and has a large change rate. The annual precipitation on the edge of Jungar basins: 150-200 mm, and about 100 mm at the center of the basin. It is less than 50 mm in the Tarim Basin, and only 20 mm in the Turpan Depression. However, because the precipitation in few areas and mountainous areas is 300-400 mm, there are also the semi-arid and semi-humid areas, which provide conditions for the formation of the steppe grasslands. The humid region with high altitude and with the precipitation more than 500 mm in the middle of Tianshan Moutains provides conditions for the formation of mountainous meadow pasture. Xinjiang is also rich in light in a large area. The annual total solar radiation in northern Xinjiang is about 130 kcal/cm and in southern Xinjiang is 145 kcal/cm. The above is conducive to the forage growth of the grasslands. Heat includes air temperature and accumulated temperature as well as frost, which have a close relationship with forage growth and the distribution of forage species. For example, tall grasses are only

Description

distributed under the conditions that the temperature is low and the accumulated temperature is low. The annual average temperature is 9-12°C in the southern Xinjiang, less than 4-9°C in the northern Xinjiang, and 10-14°C in the eastern Xinjiang. The temperature in mountainous areas is lower than that in the plain areas. For the accumulated temperature, it is the most in the eastern Xinjiang, moderate in the southern Xinjiang and least in the northern Xinjiang of 10°C. The accumulated temperature is 5300-5400°C in the eastern Xinjiang, 4000-4300°C in the southern Xinjiang, and 2500-2900°C in the northern Xinjiang. According to the rough calculation of Xinjiang Meteorological Bureau, the average latitude moves northward by one degree, and the accumulated temperature (>10C) will be reduced by 100°C, and when the altitude increases by 100 from basin to mountains, the accumulated temperature (>10°C) will decrease by 120-150°C. The frost-free period is 200-300 days in the southern Xinjiang and 140-155 days in the northern Xinjiang, but the frost-free period in mountainous areas decrease. The annual precipitation in the Northern Xinjiang accounts for 60-70 %, and the annual precipitation in the southern Xinjiang accounts for 60-70%. The snowfall and accumulated snow are still more in the northern Xinjiang and less in the southern Xinjiang. Therefore, the cold period in the northern Xinjiang is long. The thickness of the accumulated snow is generally 25-30 cm in Altay, 25 cm in Tacheng, 20 cm in Yili, and 15 cm on the northern slope of the Tianshan Mountains. The rainfall and the accumulated snow have a direct impact on the grassland and the growth of grass. From the perspective of soil and hydrogeological conditions, it is also one of the important conditions for the formation of different types of grasslands. The soil formation process and the geographical distribution law of the soil are obviously affected deeply by the strong dry climate and geology and landforms, and are consistent with the formation of grassland types. For the hydrological conditions, due to the influence of topography and climate, mountainous areas are runoff formation areas, and the plain areas are runoff loss areas, which also have a great impact on the formation of different types of grasslands. The effect of surface water and underground water in valleys and low flats provides conditions for the formation of different types of lowland medow grasslands. At present, the degradation of the grassland has already become a worldwide severe ecological problem and causes the productivity reduction of the grassland in the world by 43.0%. It is estimated that about 20% of the grassland biological yield has decreased globally. The grassland degradation area in Asia is most serious, reaching 37 million km2 , accounting for 22% of the total grassland area. Compared with the grassland degradation in the world and Asia, the degradation of grasslands in China is more serious. In the 21st century, the grassland degradation

Description

has been worsened, and 90% of the available natural grasslands in China have been degraded to varying degrees. The temperate desert grassland in Xinjiang, which accounts for one-fifth of the grassland area in northern China, is more severely degraded. The grasslands with severe degradation, desertification, and salinization have exceeded 8 million hm2 , which is increasing at a rate of 100,000 hm2 every year. The annual grass yield decreased by 20%- 4 0% in recent ten years. Because the grasslands on the northern slope of the Tianshan Mountains in Xinjiang are distributed in central and western Asia and located in the hinterland of Asia and Europe, and controlled by high-pressure anticyclones from Siberia and Mongolia, the entire area is covered by a strong temperate continental arid climate, and at the same time, it is affected by the Gurbantungut Desert, thereby forming a unique grassland type. The severe degradation of the grasslands on the northern slope of Tianshan Mountains in Xinjiang not only restricts the development of the grassland animal husbandry in Xinjiang, and directly threatens the lives of herders, but also affects the ecological balance of the region and the overall ecological security and social stability of western China. In conclusion, the problems in the prior art are as follows: The existing grassland soil degradation evaluation technology cannot provide a basis for the serious degradation of the grassland which restricts the development of the grassland animal husbandry and directly threatens the lives of herders, and cannot provide reasonable suggestions for improving or restoring the nutrients of the grassland soil.

Summary

For the problems in the prior art, the present invention provides a grassland soil degradation evaluation method. The present invention is implemented as follows: a grassland soil degradation evaluation method is provided. The grassland soil degradation evaluation method includes the following steps: Selecting an evaluation target and completing automatic identification: during completion of automatic identification of the evaluation target, acquiring an image of a grassland soil degradation geological layer by a geological image collector of a monitoring video display unit; acquiring the image of the grassland soil degradation geological layer transmitted by the geological image collector by a fuzziness evaluation module built in the monitoring video display unit, and calculating a ratio of image statistics information before and after filtering; and adjusting the fuzziness of the original image by a fuzziness adjusting module built in the

Description

monitoring video display unit and connected with the fuzziness evaluation module to obtain a final image and an image fuzziness evaluation index, wherein a method for evaluating the fuzziness of the image by using the fuzziness evaluation module and the fuzziness adjusting module includes: Step I, acquiring the image: acquiring the image of the grassland soil degradation geological layer by the geological image collector; Step II, graying the image: for convenience in extracting edges of the image, converting the color image into a gray image by using a conversion relationship between a pixel value of R, G, and B channels of the RGB image in the digital image processing and the pixel value of the gray image, wherein the formula is as follows: Gray =R*O.3+G*O.59+B*O.11;

Step III, extracting the edges of the image: using a Roberts operator edge detection technology in a digital image processing method to act on the gray image to obtain the edges of the image, wherein different detection operators have different edge detection templates, the difference of the crossed pixels is calculated as the current pixel value according to a specific template, and the used template is as follows:

E (i, j) =F (i , j) -F (i+1, j+1) |+|F (4+1,j) -F (i, j+1)| Step IV, processing the image: filtering the gray image by using a high-pass/low-pass filter to construct a reference image of a to-be-evaluated image; using a 3*3 mean filter to traverse each pixel of the image with a filter template, and placing the center of the template at the current pixel each time; and using an average value of all pixels in the template as the new value of the current pixels, wherein the template is as follows:

Step V, calculating image edge statistics information: calculating the gray information of each edge before and after the filtering respectively, wherein the statistics information of the to-be-evaluated image F before the filtering is sum-orig, the statistics information of the reference image F after the filtering is sum_ filter, and the specific calculation formulas are as follows:

Description sum _orig =w.x(F(,j)-F(Fi,j-) +F(ij)-F(i,j+I )+ IF(i, j)-F(i+i1,j)) +w2,-IF(ij)-F(i-1,j-1)1+F(ij)-F(i-1,J+1) +F(ij)-F(i+1,j-1)I+IF(i,j)-F(i+1j+1))

sum filter= w (F2(ij )-F2(i-1,j)+ F2(ij)-F2!(i,i -1) + F2(i,j)-F2(Lj+1)1+F2( ij)-F2(i+1, j) +w2x(IF2(i,)-F2(i-1,j-)+F2(,j)-F2(i-j1)+F2(i,j)-F2(i+1,j-)+ 2(isj)-F2(i+I,j+ 1))

In the formulas, wl and w2 are weight values set according to a distance to the central pixel, and W1=l, w2 =1/3; Step VI, calculating a fuzziness index of the image: using a ratio of the statistics information of the edge gray before and after the image filtering obtained in step V as the fuzziness index; and to facilitate the evaluation, taking the larger one as the denominator, taking the smaller one as the numerator, and keeping the value between (0,1); Step VII, according to a DMOS range of a best visual effect, obtaining a corresponding fuzziness index range [min, max], which is specifically: Obtaining a fuzziness adjusting range, evaluating 174 Gauss fuzzy images in LIVE 2 by using a fuzziness evaluation method in the above step, calculating the fuzziness evaluation values, then establishing a mapping relationship between the evaluation value and DMOS by using a fitting tool plot (value, DMOS), and obtaining the fuzziness evaluation range [min, max] according to the DMOS range corresponding to the best visual effect; Step VIII, adjusting the fuzziness of the image: if the fuzziness index of the image is less than min, the change of the image before and after the filtering is great and the original image is excessively sharpened according to step VI, using a low-pass filter to perform the filtering adjustment; and if the fuzziness index of the image is greater than max, the image before and after the filtering is slightly changed and the original image is excessively fuzzy, using a high-pass filter to perform the filtering adjustment, thereby achieving the best visual effect; Step IX, obtaining a final image and the fuzziness evaluation index of the image; Inputting the original data of target land: receiving the original data of the target land by a signal acquisition module built in the monitoring video display unit to obtain the final image fuzziness evaluation index information, wherein the signal acquisition module is in receiving, Firstly, using a sensing device embedded in the signal acquisition module to collect a target signal x(t) within an independent sampling period, and digitally quantifying the signal in an A/D mode; then performing the dimension reduction for the quantified signal x(i); and finally, reconstructing the dimension-reduced signal, wherein t is a sampling moment, and i is a quantified signal sequence;

Description

The dimension reduction for the quantified signal is specifically to design a filter-based compression sensing signal acquisition framework and construct a Toplitz measurement matrix LI

as follows through a difference equation k of a finite pulse h (O) -h (L-1) response filter for the quantified signal, wherein ' ' is a filter coefficient:

observing wherein bi,, bL: is regarded as a filter

coefficient; a singular value of the sub-matrix is an arithmetic root of Gramian matrix G (QF,I) ~ ( FT (FT characteristic value, all characteristic values (Ai (1-K,1+5) , i =1, -T,) of G (D F, T) are verified, so 0 F satisfies RIP, and the original

signal is reconstructed by solving an optimal problem of , that is, the original signal is reconstructed by a linear planning method, i.e. a BP algorithm;

For the collection of an actual compressed signal such as an image signal, F is changed in the following form:

If the signal has sparseness in a transformation base matrix , the original signal is

precisely reconstructed by solving the optimal problem of mn ||a||,zy= (D = Wa =Za

wherein 0 is not related to W, and 2is called CS matrix. Evaluating part of the original data of the target land; Evaluating basic characteristics of the target land: for the evaluation on the basic characteristics of the target land, evaluating various to-be-evaluated indexes by a target land basic characteristic evaluation module built in the monitoring video display unit through a frequency statistics method so as to grade the degradation risk degree of the grassland soil

Description

according to an evaluation set and to obtain a degree of membership of a factor set; and determining the evaluation membership matrix by the target land basic characteristic evaluation module: Obtaining a relative membership degree matrix of the kthfactor set:

In the formula, In the formula, R the relative membership degree matrix of thekth factor; rkij-the membership degree of j of the evaluation set where the ith factor of the kth factor set belongs; pij-frequency of the ith factor with the index graded as j by the members of the kth factor set; Constructing a fuzziness evaluation matrix:

The fuzziness evaluation matrix B can be constructed by a vector *and the matrix R of various indexes,

Calculating a comprehensive evaluation result:

The comprehensive evaluation result Zcan be solved by the fuzziness evaluation matrix B and the parameter column vector of the evaluation set;

Z=B•V The fuzziness comprehensive evaluation result can be obtained from the above formula, and then the failure risk of various factors on the grassland soil degradation can be evaluated according to the evaluation grade; and then the evaluation result is displayed by a display of the monitoring video display unit; Evaluating the homogeneity target land; Evaluating the complexity target land: acquiring the signalby a complexity target land evaluation processing module built in the monitoring video display unit, collecting data by a sensor embedded in the complexity target land evaluation processing module, and then magnifying the signal and carrying out the evaluation; extracting four basic time-domain

Description

parameters such as mean value, variance, accumulated value of the signal and peak value are extracted from each segment of signal, and judging whether possible abnormality occurs by a difference of the four parameter values of the adjacent segments of signal; if yes, downwardly executing the wavelet packet denoising; otherwise, skipping to the step of acquiring the signal; using a wavelet packet algorithm to denoise the collected signal; using the wavelet packet algorithm to perform the wavelet decomposition and reconstruction for the collected signal to obtain a single sub-band reconstructed signal; extracting eight parameters representing the signal characteristics such as the time-domain energy, time-domain peak value, frequency-domain energy, frequency-domain peak value, kurtosis coefficient, variance, frequency band and skewness coefficient from the reconstructed single sub-band signal; and using a main component analysis method to select 3 to 8 parameters capable of representing the characteristics of the complexity target land from the above parameters to form characteristic vectors, inputting these characteristic vectors into a support vector machine to perform the decision judgment, and judging whether the abnormality occurs according to the output of the support vector machine. Further, the wavelet packet denoising and the wavelet packet decomposition and reconstruction include: Signal extension: performing parabola extension for each layer of signal decomposed by the wavelet packet; Setting the signal data as x (a), x (a+) and x (a+2), wherein an expression of the

{x(af- extension operator E is

)3")-xax(oa + +

X(a±+3) - x(a±+2)-3)x(a +1I+x(a) Eliminating surplus frequency components of a single sub-band; Convoluting the extended signal with a decomposition low-pass filter ho to obtain a low-frequency coefficient, then processing with an HF-cut-IF operator, removing the surplus frequency component, then performing the downsampling to obtain a low-frequency coefficient of a next layer; convoluting the extended signal with a decomposition high-pass filter go to obtain a high-frequency coefficient, then processing with the LF-cut -IF operator, removing the surplus frequency component, and performing the downsampling to obtain a high-frequency coefficient of the next layer, wherein the HF-cut-IF operator is shown in formula (2), and the LF-cut-IFoperator is shown in formula (3);

Description

N-I N. 3N 4 4 X(kO=0,

(3) Ax(n)=Z~)

_~~~ k- - 0,1-,j1n01-,N1 In the formula (2) and the formula (3), x(n) isa coefficient of the wavelet packet ata 2 scale, and Nj indicates a length of data at the 2i scale,

A reconstruction method of the single sub-band signal includes: Performing the upsampling for the obtained high-frequency coefficient and low-frequency coefficient, then performing the convolution respectively with the high-pass reconstructed filter gi and the low-pass reconstructed filter hi, processing the obtained signal respectively with HF-cut-IF and LF-cut-IF operators to obtain the single sub-band reconstructed signal. Further, the evaluation of the basic characteristics of the target land includes: evaluating agro-ecological functions of soil fertility, a cropping system and antipollution capacity. Further, the evaluation of the complexity target land includes: evaluating an area, a boundary line and a specific target. Further, the evaluation of the agro-ecological functions includes: Determining a measurement target of soil functions and ecological quality; Formulating sufficient soil diagnosis indexes which should not exceed a target range; Selecting a rational soil diagnosis quantification grading standard; Analyzing soil ecological characteristics according to the impact of local soil formation factors on the formation and distribution law of the soil. Further, the soil ecological characteristics include water reserve, carbon reserve, available nutrient content, soil geology and agricultural chemistry.

In

Description

Further, the soil ecological characteristics also include soil restrictive factors. The soil restrictive factors include basic physical and chemical types of soil degradation, soil self-purification capacity and soil pollution. Further, evaluating the uniformity target land includes: Establishing a geographical position or coordinate of the specific land; Inputting the obtained data; Determining evaluation indexes and tables of the standard; especially formulating and selecting basic data suitable for the local soil ecological characteristics; drawing an effect graph of analysis materials; Deleting, saving, compiling, summarizing, and applying. Further, the grassland soil degradation evaluation method also includes evaluating surface power of the soil. The present invention has the following advantages and beneficial effects: The grassland soil degradation evaluation method provided by the present invention is suitable for making decisions on land adaptability in the comprehensive evaluation system of different levels under different initiation conditions. The present invention formulates the type of the agro-ecological system according to the conditions such as soil restrictive factors, soil fertility conditions, irrigation water, soil degradation and the like. The functions of the land are formulated according to the basic conditions and production characteristics of the soil. The present invention formulates a natural protection information system of the agricultural environment. The present invention provides data related to the adaptability of the natural resources and agricultural resources for determining the agricultural specialization and crop planting systems. The present invention provides reference data for evaluating soil conditions suitable for the local agricultural environment and local agro-ecological characteristics. The present invention is suitable for instructing crop species, varieties and production technologies of the local agricultural ecosystem. The present invention provides a basis for regulating and controlling the ecological agriculture and intensive agriculture combining the local ecological system and agricultural economy. The present invention provides a basis for improving the erosion of the degraded agricultural soil, thereby guaranteeing the stabilization of the agricultural ecosystem and the stability of the agricultural environment. The image evaluation of the present invention is different from the traditional evaluation method. On the basis of the structural characteristics of the to-be-evaluated image, from the perspective of relative evaluation perspective, the present invention utilizes the filter to construct

Description

the reference image of the to-be-evaluated image and calculates the ratio of the edge statistics information of the image before and after the change as the evaluation index. The present invention is simple in principle, realizes the content independence and real-time property for the evaluation of the image fuzziness, and can rapidly and accurately evaluate and compare the fuzziness between any images, thereby obtaining an accurate geological image of the grassland soil degradation, and guaranteeing the subsequent evaluation. The present invention integrates the signal acquisition, processing and evaluation to accurately evaluate the degradation information of the grassland soil. Compared with the prior art, the accuracy of the obtained signal can be increased by 6% by experiments, which can reach 97.12%.

Description of Drawings

Fig. 1 is a flow chart of a grassland soil degradation evaluation method provided by embodiments of the present invention.

Detailed Description

To make the purpose, technical solutions and advantages of the present invention more clear and understandable, the present invention is further described in detail below in combination with embodiments. It shall be understood that the specific embodiments described herein are merely used to interpret the present invention, rather than limiting the present invention. The application principle of the present invention is described in detail below in combination with the accompanying drawings. As shown in Fig. 1, a grassland soil degradation evaluation method provided by embodiments of the present invention includes: S101: an evaluation target is selected, and the automatic identification is completed; S102: original data of target land is inputted; S103: part of the original data of the target land is evaluated; S104: basic characteristics of the target land are evaluated; S105:a homogeneity target land is evaluated; S106: a complexity target land is evaluated. Further, during the completion of the automatic identification of the evaluation target, an

Description image of a grassland soil degradation geological layer is collected by a geological image collector of a monitoring video display unit; the image of the grassland soil degradation geological layer transmitted by the geological image collector is acquired by a fuzziness evaluation module built in the monitoring video display unit, and a ratio of image statistics information before and after the filtering is calculated; and the fuzziness of the original image is adjusted by a fuzziness adjusting module built in the monitoring video display unit and connected with the fuzziness evaluation module, and a final image and a fuzziness evaluation index of the image are obtained, wherein a method for evaluating the fuzziness of the image by using the fuzziness evaluation module and the fuzziness adjusting module includes: Step I, the image is acquired: the to-be-evaluated image of the grassland soil degradation geological layer is acquired by the geological image collector; Step II, the image is grayed: for convenience in extracting edges of the image, the color image is converted into a gray image by using a conversion relationship between the pixel value of R, G, and B channels of the RGB image in the digital image processing and the pixel value of the gray image. The formula is as follows:

Gray-R*0.3+G*0.59+B*0.11. Step III, the edges of the image are extracted: a Roberts operator edge detection technology in a digital image processing method is used to act on the gray image to obtain the edges of the image; different detection operators have different edge detection templates; and the difference of the crossed pixels is calculated as the current pixel value according to the specific template. The used template is as follows: E (i, j) =|IF (i , j) -F (i+1, j+1) |+|IF (i+1 , j) -F (i, j+1)|.

Step IV, the image is processed: a high-pass/low-pass filter is used to filter the gray image to construct a reference image of the to-be-evaluated image; a 3*3 mean filter is used to traverse each pixel of the image by using a filter template, and the center of the template is placed at the current pixel each time; and an average value of all pixels in the template is used as the new value of the current pixel, wherein the template is as follows:

9

Step V, edge statistics information of the image is calculated: the gray information of each edge before and after the filtering is respectively calculated; the statistics information of the to-be-evaluated image F before the filtering is sum_orig, and the statistics information of the

1A

Description

reference image F after the filtering is sum_ filter, wherein the specific calculation formulas are as follows: sum_og wi(IF(Lj) F(-1,j6F(IJ)( j-F(ij-)- F(i,j)F(,j+1)+F(j)-F(+1] j)!)

+w2-(IFitj)-F( 1j-1)I+F(,j)-F.(1lj+1)+±f(Lj)-F(i+1,jI1 F( ij)-F(i+1,,el)|} slmr= (I2(j) F2( -j)+ F2(iJ)-F2( j-l)I+IF2 j)-F2( j+ + F2 (i j- (-Lj)J) +w 2 IF2Q,)-F2(i -la1)|+|F2(iM)-F(1-1j+1I|++F2(i j)-F2(i+1 -1)+- F2|, )-F2i-L,j+)

In the formulas, wl and w2 are weight values set according to a distance to the central pixel, and wl=I, w2 =1/3; Step VI, an fuzziness index of the image is calculated: a ratio of the gray statistics information of the edge before and after the image filtering obtained in step V is used as the fuzziness index; and to facilitate the evaluation, the larger one is taken as the denominator, the smaller one is taken as the numerator, and the value is kept between (0,1); Step VII, according to a DMOS range of a best visual effect, a corresponding fuzziness index range [min, max] is obtained, which is specifically: A fuzziness adjusting range is obtained; 174 Gauss blurred images in LIVE 2 are evaluated by using a fuzziness evaluation method in the above step; the fuzziness evaluation values are calculated; then a mapping relationship between the evaluation value and DMOS is established by using a fitting tool plot (value, DMOS); and a fuzziness evaluation range [min,

max] is obtained according to the DMOS range corresponding to the vest visual effect; Step VIII, the fuzziness of the image is adjusted: if the fuzziness index of the image is less than min, the change of the image before and after the filtering is great and the original image is excessively sharpened according to step VI, then a low-pass filter is used to perform the filtering adjustment; and if the fuzziness index of the image is greater than max, the image before and after the filtering is slightly changed and the original image is excessively fuzzy, then the high-pass filter is used to perform the filtering adjustment, thereby achieving the best visual effect; Step IX, a final image and the fuzziness evaluation index of the image are obtained. Further, the original data of the target land is received by a signal acquisition module built in the monitoring video display unit to obtain the final image fuzziness evaluation index information; the signal acquisition module is in receiving, Firstly, a sensing device embedded in the signal acquisition module is used to collect a target signal x(t) within an independent sampling period, and the signal is digitally quantified in an A/D mode; then the quantified signal x(i) is subjected to dimension reduction; and finally, the dimension-reduced signal is reconstructed, wherein t is a sampling moment, and i is the

1A

Description quantified signal sequence; The dimension reduction for the quantified signal is specifically to design a filter-based compression sensing signal acquisition framework and construct a Toplitz measurement matrix

ydi) ~ik)x-k)4i=I,-,M, as follows through a difference equation k of a finite pulse response filter for the quantified signal, wherein h(0). .h(l 1) are filter coefficients:

y(i) Zbiq+l-1,i =1,• M is observed, whereinb,...,bLis regarded as a filter

coefficient; a singular value of the sub-matrix 0 FT is an arithmetic root of Gramian matrix

G (D F, T) FT D FT characteristic value; all characteristic values Xi(E (1-4+ , i=1, ---,Tof G(WF, T) are verified, so F satisfies RIP; and the original signal is reconstructed by solving an

optimal problem of ,that is, the original signal is reconstructed by a linear planning method, i.e. a BP algorithm;

For the collection of an actual compressed signal such as an image signal, OF is changed to the following form:

h.I

If the signal has sparseness in a transformation base matrix Uthe original signal is

precisely reconstructed by solving the optimal problem ofmna v, v = ,=- 'a=

wherein C is not related to W, and Eis called CS matrix. Further, the signal is acquired by a complexity target land evaluation processing module built in the monitoring video display unit; data is collected by a sensor embedded in the complexity target land evaluation processing module, and then the signal is magnified and evaluated; four basic time-domain parameters such as mean value, variance, accumulated value of the signal and peak value are extracted from each segment of signal, and whether possible

1s

Description

abnormality occurs is judged by a difference of the four parameter values of the adjacent segments of signals; if yes, wavelet packet denoising is executed downwardly; otherwise, the step of acquiring the signal is executed; a wavelet packet algorithm is used to denoise the collected signal; the wavelet packet algorithm is used to perform the wavelet decomposition and reconstruction for the collected signal to obtain a single sub-band reconstructed signal; eight parameters representing the signal characteristics such as the time-domain energy, time-domain peak value, frequency-domain energy, frequency-domain peak value, kurtosis coefficient, variance, frequency band and skewness coefficient are extracted from the reconstructed single sub-band signal; a main component analysis method is used to select 3 to 8 parameters capable of representing the characteristics of the complexity target land from the above parameters to form characteristic vectors; these characteristic vectors are inputted into a support vector machine to perform the decision judgment, and whether the abnormality occurs is judged according to the output of the support vector machine. Further, the wavelet packet denoising and the wavelet packet decomposition and reconstruction include: Signal extension: each layer signal decomposed by the wavelet packet is subjected to the parabola extension; The signal data is set as x (a), x (a+1) and x (a+2), wherein an expression of the extension operator E is:

{x(a-1)=3x(a)-3x(a+1)+xa+2) x(a+3)- 3x(a+2)-3x(a+1)+x(a) Surplus frequency components of a single sub-band are eliminated; The extended signal is convoluted with a decomposition low-pass filter ho to obtain a low-frequency coefficient, then the signal is processed with an HF-cut-IF operator, the surplus frequency component is removed, and then the downsampling is performed to obtain a low-frequency coefficient of a next layer; the extended signal is convoluted with a decomposition high-pass filter go to obtain a high-frequency coefficient, then the signal is processed with the LF-cut-IF operator, the surplus frequency component is removed, and the downsampling is performed to obtain a high-frequency coefficient of the next layer, wherein the operator HF-cut-IF is shown in formula (2), and the LF-cut-IFoperator is shown in formula (3);

16C

Description

X(k) - x(n)WA r0 k<~i 4 4 ' fL; N X(k)-0,

x(f) - x(k)W-", (2

X(k)= x4n)WyL k<J 4 4 X(k)-0, N, -1 (a)

In the formula (2) and the formula (3), x(n) is a coefficient of the wavelet packet at a 2i

scale, and Nj indicates a length of data at the 2i scale,W= k=,1,,Nj1;n=,1,,Nj1

A reconstruction method of the single sub-band signal includes: The obtained high-frequency coefficient and low-frequency coefficient are subjected to upsampling, then convoluted respectively with the high-pass reconstructed filter gi and the low-pass reconstructed filter hi, and the obtained signal is processed respectively with HF-cut-IF and LF-cut-IF operators to obtain the single sub-band reconstructed signal. Further, for the evaluation on the basic characteristics of the target land, various to-be-evaluated indexes are evaluated by a target land basic characteristic evaluation module built in the monitoring video display unit through a frequency statistics method so as to grade the degradation risk degree of the grassland soil according to an evaluation set and to obtain a degree of membership of a factor set; and the evaluation membership matrix is determined by the target land basic characteristic evaluation module: A relative membership degree matrix of the kth factor set is obtained:

In the formula, In the formula, Rk-the relative membership degree matrix of the kth factor;

rkij-the membership degree of j of the evaluation set where the ith factor of the kth factor belongs;

Description

pkij-frequency of the ith factor with the index graded as j by the members of the kth factor set;

A fuzziness evaluation matrix is constructed:

The fuzziness evaluation matrix B can be constructed by a vector and matrix R of various indexes,

B -We A comprehensive evaluation result is calculated:

The comprehensive evaluation result Z can be solved by the fuzziness evaluation matrix B and the parameter column vector of the evaluation set;

Z=B•V The fuzziness comprehensive evaluation result can be obtained from the above formula, and then the failure risk of various factors on the grassland soil degradation can be evaluated; and then the evaluation result is displayed by a display of the monitoring video display unit. Further, the evaluation of the basic characteristics of the target land includes: agro-ecological functions of soil fertility, a cropping system, and antipollution capacity are evaluated. Further, the evaluation of the complexity target land includes: an area, a boundary line and a specific target are evaluated. The evaluation of the agro-ecological functions includes: a measurement target of soil functions and ecological quality is determined; sufficient soil diagnosis indexes are formulated, which should not exceed a target range; a rational soil diagnosis quantification grading standard is selected; and soil ecological characteristics are analyzed according to the impact of local soil formation factors on the formation and distribution law of the soil. The soil ecological characteristics include water reserve, carbon reserve, available nutrient content, soil geology and agricultural chemistry. The soil ecological characteristics also include soil restrictive factors. The soil restrictive factors includes basic physical and chemical types of soil degradation, soil self-purification capacity and soil pollution. The evaluation of the homogeneity target land includes: a geographical position or coordinate of the specific land is established; the obtained data is inputted;

Description

evaluation indexes and tables of the standard are determined; basic data suitable for the local soil ecological characteristics are formulated; an effect graph of analysis materials is drawn; and deleting, saving, compiling, summarizing, and applying are carried out. The grassland soil degradation evaluation method also includes evaluating surface power of the soil. The present invention is further described below in combination with specific embodiments. 1. Influence of types of artificial power on change of soil quality One of the realistic factors of the soil surface power is human activity power, which has great influence on the soil surface layers. Since the 20th century, many scientists have continuously studied the influence of geological and ecological factors on the soil and agricultural environment from different perspectives, and have proposed a process of influencing the development of the soil surface layers and the improvement or deterioration of various ecological characteristics. Importance of human factors. The main performance is the change of soil properties and components, change of soil bulk density and atmospheric composition. The effect on the aspects such as the change rate of climate is great. The degradation or deterioration of major soil under the natural environmental condition is also caused by the incorrect human technical services. Therefore, the influence of the human factor on the soil surface layers may become one of general phenomena that can be seen anywhere on the earth. A specific way for the human factors to cause the soil degradation is as follows: The violate reclamation accelerates the change of the soil surface layers, which causes the profound change of the ecological characteristics of soil sections, soil structure and other physical and chemical characteristics. Especially the blind application of various chemical and organic fertilizers causes the imbalance of the soil nutrients. The agricultural physical degradation of the soil worsens the water-to-air ratio, destroys the soil structure and increases the bulk density. The agricultural technological degradation of the soil worsens the physical and mechanical properties of the topsoil. The agricultural chemical loss is the imbalance of other nutrients in the soil caused by the increased application of a certain nutrient. The reduction of the number of biogroups or the death of organisms in the soil and the pollution of soil environment caused by the application of pesticides are results of degradation of soil ecological characteristics caused by the human activities.

Description

In general, the unreasonable utilization of soil may accelerate the degradation of the soil and especially worsens the degradation of the surface soil, causing the rapid reduction of soil fertility. Another major factor of the human activities is irrigation, which also has great impact on the soil properties. The irrigation not only can destroy the soil sections and the water-to-air ratio of the soil, but also may influence the amount of pH, Eh, Ca2+, K+ and N03-N of the soil. Therefore, the unreasonable irrigation may accelerate the loss of water and soil. The destruction of the soil structure makes the soil compact, and causes the leaching loss of the base ions and acidification of the soil; and on the contrary, it can cause the soil salinization and alkalization. 2. Conventional agricultural evaluation factors for comprehensive evaluation of soil agro-ecology: There are six agro-ecological functions as follows: (1) A function for providing nutrients needed by organisms It is an agro-ecological function of soil nutrients capable of being absorbed and utilized by plants, especially their supply characteristics and productivity. (2) Coordination of soil moisture: one of main factors determining the effectiveness of the soil moisture is the physical and mechanical properties of the soil, and is the most important physical properties of the soil. (3) Dependence on physical and mechanical properties of the soil. (4) The morphological properties of the soil depend on the stability of soil fertility factors. (5) Soil colloid determines the self-cleaning capacity of the soil for the pest and disease control pesticides. (6) The soil geo-chemical function determines the self-cleaning capacity of the soil for metal pollutants. 3. Modern soil ecological evaluation At present, there are five kinds of soil ecological evaluation as follows: (1) Measurement targets, i.e. soil functions and ecological quality are determined. (2) Sufficient soil diagnosis indexes are formulated, and should not exceed the range (target range). (3) A rational soil diagnosis quantification grading standard is selected. (4) Ecological characteristics of the soil are evaluated according to a model. (5) When soil ecological characteristics are described, attention should be paid to the impact of local soil formation factors on the formation and distribution law of the soil. 4. The application of a grassland soil degradation evaluation system (PACKA3) provided by embodiments of the present invention should follow the follow six principles:

) A framework of the system; @ an ecological method is used to explain results; @

functions (basic or partial agro-ecological functions) of the evaluation indexes; @ the

Description

agro-ecological system of the target land is evaluated according to the national evaluation indexes; @ flexibility adaptive to the natural and technological conditions of the target land; and

@ determination of restrictive factors of the target land, i.e. determination of restrictive factors

influencing the local production and agro-ecology. The automatic comprehensive evaluation of the soil ecological characteristics requires the following three procedures:

0 Determination of basic diagnosis indexes. @ Analysis on agro-ecological

requirements for the local main crops. @ The local soil diagnosis index and agro-ecological

characteristics and cultivation technology for the production of crops are estimated according to the standard in report forms. The estimation steps are as follows:

( An evaluation target is selected, and the automatic identification is completed;@

Original data of a target land is inputted; @ Part of the original data of the target land is

evaluated; @ The basic characteristics of the target land are evaluated (agro-ecological functions

of soil fertility, a cropping system, and antipollution capacity ). @ The uniformity target land is

evaluated; @ The complexity target land is evaluated, including an area, a boundary line and a

specific target. Automatic comprehensive evaluation on the soil ecological characteristics provided by embodiments of the present invention includes:

( Automatic comprehensive evaluation model principle of the soil ecological

characteristics at a key point is established. @ Main soil diagnosis indexes are established and

determined. @ Basic standards, tables and calculation methods for the automatic comprehensive

evaluation model of soil ecological characteristics are established and determined. @ The soil

diagnosis method is used to accurately and flexibly describe and analyze the composition of the model to solve the analysis method of test data. @ An application model range is selected

according to specific local situation. For the specific land:

( A geographical position (coordinate) of the specific land is established. @ The

obtained data is inputted. @ The evaluation indexes, tables and calculation methods of the

standard are determined and analyzed. The basic data suitable for the local soil ecological characteristics is formulated. @ An effect graph of analysis materials is drawn. @ Deleting,

Description

saving, compiling, summarizing, and applying are carried out. For a specific land evaluation model:

Including 0 basic topological soil layer maps (maps of 1:100.000 and 1:200.000);@

map showing current utilization status of the land: Specifically, the map with the area of 1000 ha; @ small and medium-sized geological and geomorphological map; @ soil formation map,

including soil types and agricultural utilization thereof; @ agro-climatic map; @ small and

micro soil distribution map; The following maps can be drawn according to the evaluation data of PACKA3: The maps of soil agro-ecological characteristics include maps of water reserve, carbon reserve, available nutrient content, soil geology and agricultural chemistry; maps of soil restrictive factors include basic physical and chemical types of soil degradation, soil self-cleaning capacity and soil pollution, such as hardening. The application principle of the present invention is further described below in combination with specific embodiments. In a pasture ecological system, the soil is used as the substrate and environment for the growth of plants. The physical and chemical properties of soil have the most profound mechanism for the dynamics of plant communities. Soil pH value, organic matter content and the content of other nutritive elements show different degrees of dependence on characteristics of plant communities. The characteristics of plant communities also have significant effect on the decomposition of soil organic matters and soil nutrient dynamics. The analysis made by Yao et al on the change of the soil physical and chemical properties of the degraded grassland in 1982 and 2003 showed that compared with that 21 years ago, the soil pH value of the grassland had increased, the water retention capacity of the soil decreased by 32.5%, and the content of organic matters in the soil decreased by 5.5%. Although the content of total phosphorus and total nitrogen in soil was relatively stable, they were changed to certain degree. This result was consistent with the result of the present invention. However, the degradation of the above indexes in the soil of Xinjiang pasture is relatively serious. The existing analysis shows that the excessive trampling and eating by domestic animals may cause the drought of the grassland, worsening of the physical and chemical properties of the soil and the reduction of fertility. According to the report of Wang on Leymus chinensis grassland, with the increase of herding intensity, the soil moisture and organic matter content of the grassland decrease, and the bulk density and pH value of the soil increase gradually. With the increasing trampling impact of animals, the distribution spatial layout of soil pores is changed, the total porosity of the soil

Description

decreases, and the bulk density of the soil increases. According to the reports about the agro-chemical properties and microbial flora of the grasslands with three different degradation degrees in the prior art, after the degradation of the grassland, the soil fertility and the number and types of soil microorganisms decrease with the increase of degradation degree. In the aspect of degradation characteristics of desert grasslands, Zhu et al and Lin et al investigated the main economic characteristics of Yili Seriphidium transiliense desert grassland at different degradation stages. The investigation results showed that a series of changes occurred in the plant species composition, community height, coverage, productivity and nutritional value of the grassland from the undegraded stage to the severely degraded and extremely degraded stage. The plant species decreased; the height, coverage and productivity decreased; and the feeding value was in a deterioration process, which were reflected on poor palatability of the forage for domestic animals, low utilization rate and changes from good pasture to medium-low-grade pasture. In the aspect of desert grassland degradation monitoring, An, Li, Huang, Hou, Jin and Aikebaier used different modes to analyze the degradation ecological adaptability of the desert grassland in Xinjiang and established their own detection models. The results showed that under the long-term high grazing pressure, the diversity and persistence of Seriphidium transiliense sericulture shrubs on the northern slope of Tianshan Mountains in Xinjiang responded to the high grazing pressure by reducing edible forage species and plant quantity, shortening stems, reducing shrub density, increasing the shrub diameter and changing the growth way of branches. The content of organic matters is an important indicator for measuring the fertility of the grassland soil. Among various nutrient elements, nitrogen, phosphorus and potassium are the nutrient elements with more demand amount for the plants and more loss taken away during the harvesting of the plants. CEC basically represents the amount of nutrients that are possibly kept by the soil. It is a main source of the soil cushioning performance and an important evidence of the improved soil and rational fertilization. The pH value of the soil also has great influence on nutrient availability. The soil humic acid contains some elements needed by the growth and development of the plants, which can improve soil and increase fertility. The humic acid contains carbon and nitrogen slightly more than fulvic acid, which is the main component of soil humus. The humic acid plays an important role in the formation of soil structure. The fulvic acid usually contains 70 or more minerals and trace elements, and has good solubility and fluidity. It is calculated that since the second national soil survey in 1983, the change law of the physical and chemical properties of the grassland soil in Yili area has been obtained. The content of

Description

organic matters has decreased significantly in 30 years, from 67.17g/kg 30 years ago to the current 50.6g/kg, with the rate of change of -24.65%, which has reached the medium-level degradation. In the past 30 years, the total amount of N, P, K, alkali-hydrolyzable nitrogen and quick-acting P and K content in the typical rubbing Australian soil in Yili of Xinjiang have also changed significantly. The contents decreased by 39.40%, 33.17%, 18.00%, 87.56%, 31.33% and -40.19% respectively, and the changes in total nitrogen and alkali-hydrolyzable nitrogen content decreased particularly significantly. The C/N and pH value of the soil increased by 21.16% and 3.93% respectively, while the cation exchange quantity changed greatly, which decreased by 60.93%. It can be seen that the soil nutrient degradation degree of Xinjiang grasslands is the highest at present, and it is necessary to find ways to improve or restore the soil nutrient status of the grasslands. The content of humic acid and fulvic acid in the organic matters decreased by 36.90% and 32.20% respectively, and CH/CF decreased by 6.45%, which indicated that not only the content of organic matters had a decrease trend, but also the quality of humus had a decrease trend. Soil texture is one of the physical properties of the soil. The soil texture is an important basis for drawing up utilization, management and improvement measures of the soil. The bulk density of the soil indicates the tightness and porosity of the soil, and reflects the water permeability, ventilation and resistance to the growth of plant roots. The porosity of the soil directly affects the soil moisture and ventilation, thereby affecting the growth of plants on the ground. The field water retention capacity has been regarded as the highest soil moisture capable of being stably kept by the soil for a long time and has an important significance for the growth of vegetation. It can be seen from Table 1 that the content of sand particles and powder particles of the soil has increased in different degrees in the past 30 years; and especially the content of sand particles has increased obviously, reaching 72.27% and 12.03% respectively, which has reached the degree of moderate-to-slight-severe degradation, indicating that the water and fertilizer retention capacity of the soil has been greatly reduced. However, the content of clay particles of the soil has decreased, and its change rate is -56.25% in about 30 years. The bulk density and the specific gravity of the soil have both increased. The bulk density has changed greatly, with the change rate of 45.70%. The change rate of the specific gravity is 7.30%. The total porosity has decreased by -19.03%, and the field water retention capacity has decreased by 8.42%.

Description Field water Powder Clay particles Specific retention Index Sand particles particles (<0.005%) Bulk density gravity Total capacity (>0.05%) (0.05-0.005 %) (mg/rn 3) (mg/n 3 ) porosity (%) (g /k g) 30 years ago 19.36 45.93 34.70 0.86 2.48 65.45 195.00 30 years 33,35 51.46 15.18 1.25 2.66 52.99 178.59 later

Change rate (%) 72.27 12.03 -56.25. 45.70 7.30 -19.03 -8.42

Table 1 Change rate of soil physical and mechanical properties of typical pastures in Yili of Xinjiang Table 2 shows temporary indexes for evaluating the degradation of soil ecological characteristics determined by experts such as Wa Sinov I. I, professor, doctoral supervisor and director of the Department of Soil Ecology, Myanov Agricultural University, Russia, and Aikebaier Ilahon, professor and doctoral supervisor of the Department of Soil Agrochemistry, College of Grass Ring, Xinjiang Agricultural University, through field investigation and indoor analysis on the grassland soil in both Russia and China. The above results are evaluated according to the indexes.

It- UdmAe Sihy t dga Extemoy graded med y

Increase of bulk density compared with the original soil,% <t1 11_20 21-10 1-40 >40 Goit contentVo <5 6...15 16-35 3k,.70 >70 Deceas e of orgac matte content compared with the original <10 11-25 26 35 36-.45 > 45 soil

Decreas e of orgame matte res erve compared with the original < 11,20 21-40 41-80 > M soil

Change rate of pH <Q025 025,50 05-0 0,75-10 >1,0 Decreas e of total nitrogen compared with the original soil 10 1-10 11.,30 31-40 >40 Decreas e of alkali hydrolyzed nitrogen compared with the oringasoil < se.20 21-40 41-80 >go Decrease of total phosphorus compared with the original soil <10 114.20 21-30 3...40 540 Decrese of quick-acting phosphorus compared with the originalsoil <1 11-20 21-40 41--80 >8-0 Decrease of total potassium compared with the orginalsoil < .1-1 21-30 31-4-0 >40 Decrease of quick-acting potassium compared with the original soil <10 11.110 2L...10 31-40 >40 Decrease ofcation exchange content compared with the oriinalsoil <10 11..-20 21-40 41...80 >80 Decrease orincrease ofparticles comparedwiththe originalsoil I--20 21-30 1LA4 >40 Decrease or increas e of field water retention capacity compared with <10 11 21 AO .4 >40 the original s oil

Decrease orincrease of specific gravity compared with the original < 11-20 21-30 3L-40 >40 Decrease of total nitrogen compared with the originalsoil < 10 so2 21-30 n1 -4 Decrease of carbon-to-nitrogen ratio compared with the originalsoil < 1 21-30 31 ..40 >40 Decrease of huic acid compared with the origalsoil to 11-20 2L..30- 31..;4 >41 Decrease of fulvic acid compared with the oginasoil <10 11-.20 21..30 e.4 >40 Decrease ofhunmic acid/fulvic acid ratio comapred with the original soil 05 0.25-050 01 -075 0.75.10 > A.

1)5

Description

The above only describes preferred embodiments of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitution and improvements made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

-

Claims (9)

1. Claims 1. A grassland soil degradation evaluation method, comprising the following steps: selecting an evaluation target and completing automatic identification: during completion of the automatic identification of the evaluation target, acquiring an image of a grassland soil degradation geological layer by a geological image collector of a monitoring video display unit; acquiring the image of the grassland soil degradation geological layer transmitted by the geological image collector by a fuzziness evaluation module built in the monitoring video display unit, and calculating a ratio of image statistics information before and after filtering; and adjusting the fuzziness of the original image by a fuzziness adjusting module built in the monitoring video display unit and connected with the fuzziness evaluation module to obtain a final image and an image fuzziness evaluation index, wherein a method for evaluating the fuzziness of the image by using the fuzziness evaluation module and the fuzziness adjusting module comprises: step I, acquiring the image: acquiring the image of the grassland soil degradation geological layer by the geological image collector; step II, graying the image: for convenience in extracting edges of the image, converting the color image into a gray image by using a conversion relationship between a pixel value of R, G, and B channels of the RGB image in the digital image processing and the pixel value of the gray image, wherein the formula is as follows: Gray=R*0.3+G*0.59+B*0.11;

   step III, extracting the edges of the image: using a Roberts operator edge detection technology in a digital image processing method to act on the gray image to obtain the edges of the image, wherein different detection operators have different edge detection templates, the difference of the crossed pixels is calculated as the current pixel value according to a specific template, and the used template is as follows:

   E (i, j) =F (i, j) -F (i+1, j+1) |+|F (4+1,j) -F (i, j+1)| step IV, processing the image: filtering the gray image by using a high-pass/low-pass filter to construct a reference image of a to-be-evaluated image; using a 3*3 mean filter to traverse each pixel of the image with a filter template, and placing the center of the template at the current pixel each time; and using an average value of all pixels in the template as the new value of the current pixels, wherein the template is as follows:

   X

   Claims step V, calculating image edge statistics information: calculating the gray information of each edge before and after the filtering respectively, wherein the statistics information of the to-be-evaluated image F before the filtering is sumorig, the statistics information of the reference image F after the filtering is sum_ filter, and the specific calculation formulas are as follows:

   sum_orig=wlx(F(i,j)-F(i-1,j)+IF(ij)-F(i,j-)|-IF(i,j)-F(i,j+)+F(i,j)-F(i+1j ))

   um _filer - wb, j+-F2(i-1,j) +F2(ij)-F2(i,j-1)+1F2(i,j)-F2(i,j1)+ -2(i F2(i, j)- F2(i+1.,j)

   +fw2x(1F2(i,j)-F2(i-1j-1)+1F2(i,j)-F2(i-1,j+-1 +IF2(i,j)-2(i+i1,j-1)+ F2(i,j)-F2(i+1,j+1))

   in the formulas, wl and w2 are weight values set according to a distance to the central pixel, and W1=l, w2 =1/3; step VI, calculating a fuzziness index of the image: using a ratio of the statistics information of the edge gray before and after the image filtering obtained in step V as the fuzziness index; and to facilitate the evaluation, taking the larger one as the denominator, taking the smaller one as the numerator, and keeping the value between (0,1); step VII, according to a DMOS range of a best visual effect, obtaining a corresponding fuzziness index range [min, max], which is specifically: obtaining a fuzziness adjusting range, evaluating 174 Gauss fuzzy images in LIVE 2 by using a fuzziness evaluation method in the above step, calculating the fuzziness evaluation values, then establishing a mapping relationship between the evaluation value and DMOS by using a fitting tool plot (value, DMOS), and obtaining the fuzziness evaluation range [min, max] according to the DMOS range corresponding to the best visual effect; step VIII, adjusting the fuzziness of the image: if the fuzziness index of the image is less than min, the change of the image before and after the filtering is great and the original image is excessively sharpened according to step VI, using a low-pass filter to perform the filtering adjustment; and if the fuzziness index of the image is greater than max, the image before and after the filtering is slightly changed and the original image is excessively fuzzy, using a high-pass filter to perform the filtering adjustment, thereby achieving the best visual effect; step IX, obtaining a final image and the fuzziness evaluation index of the image; inputting the original data of target land: receiving the original data of the target land by a signal acquisition module built in the monitoring video display unit to obtain the final image

   Claims

   fuzziness evaluation index information, wherein the signal acquisition module is in receiving, firstly, using a sensing device embedded in the signal acquisition module to collect a target signal x(t) within an independent sampling period, and digitally quantifying the signal in an A/D mode; then performing the dimension reduction for the quantified signal x(i); and finally, reconstructing the dimension-reduced signal, wherein t is a sampling moment, and i is a quantified signal sequence; the dimension reduction for the quantified signal is specifically to design a filter-based compression sensing signal acquisition framework and construct a Toplitz measurement matrix L-I pi)= hi'lx, k) , i =lI

   , as follows through a difference equation of a finite pulse response

   filter for the quantified signal, whereinh(0) (L-)is a filter coefficient:

   h, b, 11

   observing wherein bi,, bL' is regarded as a filter

   coefficient; a singular value of the sub-matrix is an arithmetic root of Gramian matrix G ((DF, T = ' FT (l characteristic value, all characteristic values

   (AiE-(1-,i+6),i1,-,T,) OfG(DFT) areverified,so OFsatisfies RIP, and the original

   signal is reconstructed by solving an optimal problem of that is, the original signal is reconstructed by a linear planning method, i.e. a BP algorithm;

   for the collection of an actual compressed signal such as an image signal, F is changed in the following form:

   Claims

   if the signal has sparseness in a transformation base matrix V, the original signal is

   precisely reconstructed by solving the optimal problem of

   wherein 0 is not related to W, and is called CS matrix; evaluating part of the original data of the target land; evaluating basic characteristics of the target land: for the evaluation on the basic characteristics of the target land, evaluating various to-be-evaluated indexes by a target land basic characteristic evaluation module built in the monitoring video display unit through a frequency statistics method so as to grade the degradation risk degree of the grassland soil according to an evaluation set and to obtain a degree of membership of a factor set; and determining the evaluation membership matrix by the target land basic characteristic evaluation module: obtaining a relative membership degree matrix of the kth factor set:

   Rk=y

   in the formula, J

   in the formula, Pl the relative membership degree matrix of the kth factor; rkij-the membership degree of j of the evaluation set where the ith factor of the kth factor set belongs; pij-frequency of the ith factor with the index graded as j by the members of the kth factor set; constructing a fuzziness evaluation matrix:

   the fuzziness evaluation matrix B can be constructed by a vector .and the matrix R of various indexes,

   B= .R

   calculating a comprehensive evaluation result:

   the comprehensive evaluation result Z can be solved by the fuzziness evaluation matrix B and the parameter column vector of the evaluation set;

   Z=B•V

   Claims the fuzziness comprehensive evaluation result can be obtained from the above formula, and then the failure risk of various factors on the grassland soil degradation can be evaluated according to the evaluation grade; and then the evaluation result is displayed by a display of the monitoring video display unit; evaluating the homogeneity target land; evaluating the complexity target land: acquiring the signal by a complexity target land evaluation processing module built in the monitoring video display unit, collecting data by a sensor embedded in the complexity target land evaluation processing module, and then magnifying the signal and carrying out the evaluation; extracting four basic time-domain parameters such as mean value, variance, accumulated value of the signal and peak value are extracted from each segment of signal, and judging whether possible abnormality occurs by a difference of the four parameter values of the adjacent segments of signal; if yes, downwardly executing the wavelet packet denoising; otherwise, skipping to the step of acquiring the signal; using a wavelet packet algorithm to denoise the collected signal; using the wavelet packet algorithm to perform the wavelet decomposition and reconstruction for the collected signal to obtain a single sub-band reconstructed signal; extracting eight parameters representing the signal characteristics such as the time-domain energy, time-domain peak value, frequency-domain energy, frequency-domain peak value, kurtosis coefficient, variance, frequency band and skewness coefficient from the reconstructed single sub-band signal; and using a main component analysis method to select 3 to 8 parameters capable of representing the characteristics of the complexity target land from the above parameters to form characteristic vectors, inputting the characteristic vectors into a support vector machine to perform the decision judgment, and judging whether the abnormality occurs according to the output of the support vector machine.
2. 2. The grassland soil degradation evaluation method according to claim 1, wherein the wavelet packet denoising and the wavelet packet decomposition and reconstruction comprise: signal extension: performing parabola extension for each layer of signal decomposed by the wavelet packet; setting the signal data as x (a), x (a+) and x (a+2), wherein an expression of the extension operator E is

   {x(a -1)3x(a)-3x(a +1)x(a 2) x(a +3)= 3x(a + 2)-3x(a +1)+x(a) eliminating surplus frequency components of a single sub-band; convoluting the extended signal with a decomposition low-pass filter ho to obtain a

   Claims low-frequency coefficient, then processing with an HF-cut-IF operator, removing the surplus frequency component, then performing the downsampling to obtain a low-frequency coefficient of a next layer; convoluting the extended signal with a decomposition high-pass filter go to obtain a high-frequency coefficient, then processing with the LF-cut -IF operator, removing the surplus frequency component, and performing the downsampling to obtain a high-frequency coefficient of the next layer, wherein the HF-cut-IF operator is shown in formula (2), and the LF-cut-IFoperator is shown in formula (3); -1 N 3N X 'c) (n)U -Ofc :2 k N1 X~k) )4 O,

   N~i (2) x(n) - x(k)W

   NN.

   4 4 X k) - ,

   N-I (3)

   kAO

   in the formula (2) and the formula (3), x (n) is a coefficient of the wavelet packet at a 2i scale, and Nj indicates a length of data at the 2i scale,

   _- -" k=0o,1,--Nj-1I n = 0,1,I -Nj-1

   a reconstruction method of the single sub-band signal comprises: performing the upsampling for the obtained high-frequency coefficient and low-frequency coefficient, then performing the convolution respectively with the high-pass reconstructed filter gi and the low-pass reconstructed filter hi, processing the obtained signal respectively with HF-cut-IF and LF-cut-IF operators to obtain the single sub-band reconstructed signal.
3. 3. The grassland soil degradation evaluation method according to claim 1, wherein the evaluation of the basic characteristics of the target land comprises: evaluating agro-ecological functions of soil fertility, a cropping system and antipollution capacity.
4. 4. The grassland soil degradation evaluation method according to claim 1, wherein the evaluation of the complexity target land comprises: evaluating an area, a boundary line and a specific target.
5. 5. The grassland soil degradation evaluation method according to claim 3, wherein

   Claims the evaluation of the agro-ecological functions comprises: determining a measurement target of soil functions and ecological quality; formulating sufficient soil diagnosis indexes which should not exceed a target range; selecting a rational soil diagnosis quantification grading standard; analyzing soil ecological characteristics according to the impact of local soil formation factors on the formation and distribution law of the soil.
6. 6. The grassland soil degradation evaluation method according to claim 5, wherein the soil ecological characteristics comprise water reserve, carbon reserve, available nutrient content, soil geology and agricultural chemistry.
7. 7. The grassland soil degradation evaluation method according to claim 5, wherein the soil ecological characteristics also comprise soil restrictive factors; and the soil restrictive factors comprise basic physical and chemical types of soil degradation, soil self-purification capacity and soil pollution.
8. 8. The grassland soil degradation evaluation method according to claim 1, wherein evaluating the uniformity target land comprises: establishing a geographical position or coordinate of the specific land; inputting the obtained data; determining evaluation indexes and tables of the standard; especially formulating and selecting basic data suitable for the local soil ecological characteristics; drawing an effect graph of analysis materials; deleting, saving, compiling, summarizing, and applying.
9. 9. The grassland soil degradation evaluation method according to claim 1, wherein the grassland soil degradation evaluation method also comprises evaluating surface power of the soil.