CN111781163A - A method for eliminating the influence of soil particle size on the detection of soil parameters in discrete near-infrared bands - Google Patents

Abstract

The embodiment of the invention provides a method for eliminating the influence of soil granularity on the detection of soil parameters in a discrete near-infrared band, wherein the method comprises the following steps: determining the original absorbance of the soil to be detected in a plurality of detection wave bands under near infrared spectrum scanning; determining a granularity correction coefficient of the soil to be detected based on the original absorbance of the soil to be detected in the characteristic wave band; the characteristic wave band is determined based on the original absorbance of samples of a plurality of preset wave bands of soil samples under a plurality of soil granularity under near infrared spectrum scanning; and correcting the original absorbance of each detection wave band of the soil to be detected based on the granularity correction coefficient of the soil to be detected, and determining the soil parameters of the soil to be detected based on the correction result. The method provided by the embodiment of the invention reduces the interference of the soil granularity on the discrete near infrared band, and improves the soil parameter detection precision.

Description

A method for eliminating the influence of soil particle size on the detection of soil parameters in discrete near-infrared bands

technical field

The invention relates to the technical field of spectral detection, in particular to a method for eliminating the influence of soil particle size on soil parameters detected in discrete near-infrared bands.

Background technique

The detection of soil parameters based on discrete near-infrared bands can not only effectively reduce the detection cost, but also the established soil parameter prediction models have achieved good prediction results. However, soil parameter detection based on discrete near-infrared bands faces serious interference caused by soil particle size.

In the prior art, the soil particle size interference is usually eliminated by sieving the soil after grinding, but this method is time-consuming and labor-intensive, and cannot be applied to the soil particle size interference elimination when using discrete near-infrared wavebands for online soil detection. In addition, the differential methods commonly used for continuous near-infrared spectroscopy are not suitable for discrete near-infrared bands.

SUMMARY OF THE INVENTION

The embodiment of the present invention provides a method for eliminating the influence of soil particle size on soil parameters detected by discrete near-infrared bands, so as to solve the serious interference caused by soil particle size when soil parameters are detected based on discrete near-infrared bands in the prior art. problem of poor accuracy.

In a first aspect, an embodiment of the present invention provides a method for eliminating the influence of soil particle size on soil parameters detected in discrete near-infrared bands, including:

Determine the original absorbance of multiple detection bands of the soil to be detected under near-infrared spectral scanning;

Determine the particle size correction coefficient of the soil to be detected based on the original absorbance of the soil to be detected in the characteristic waveband; the characteristic waveband is a plurality of preset wavebands based on the near-infrared spectrum scanning of soil samples with multiple soil particle sizes The original absorbance of the sample is determined;

Based on the particle size correction coefficient of the soil to be detected, the original absorbance of each detection band of the soil to be detected is corrected, and the soil parameters of the soil to be detected are determined based on the correction result.

Optionally, determining the particle size correction coefficient of the soil to be detected based on the original absorbance of the soil to be detected in the characteristic waveband, including:

Based on the original absorbance of the soil to be detected in the characteristic waveband, determine an absorbance ratio used to characterize the soil particle size of the soil to be detected;

Based on the absorbance ratio of the soil to be detected and the absorbance ratio of the reference soil particle size, the particle size correction coefficient of the soil to be detected is determined.

Optionally, the absorbance ratio of the reference soil particle size is determined based on the original sample absorbance of the soil sample under the near-infrared spectrum scanning of the soil sample under the reference soil particle size.

Optionally, the characteristic band is obtained based on the following method:

performing near-infrared spectrum scanning on the soil samples under the plurality of soil particle sizes, and determining the original sample absorbance of the soil samples in a plurality of preset wavelength bands;

Determine the standard deviation value of the soil sample under each soil particle size, and the comprehensive standard deviation value based on the original absorbance of the sample in multiple preset bands of each soil sample;

The characteristic band is determined based on the standard deviation value of the soil sample at each soil particle size, and the integrated standard deviation value.

Optionally, the characteristic band is determined based on the standard deviation value of the soil sample under each soil particle size and the comprehensive standard deviation value, and further includes:

Determine the sample absorbance ratio of the soil sample based on the original sample absorbance of the characteristic waveband of the soil sample;

Based on the sample absorbance ratio and soil particle size of the soil sample, a soil particle size classification model is established to verify the ability of the sample absorbance ratio to characterize soil particle size.

Optionally, the characteristic wavelength bands are 1361 nm and 1870 nm.

Optionally, the soil sample includes 4 particle size grades and 6 total nitrogen concentration gradients, the 4 particle size grades are 0.2 mm, 0.45 mm, 0.9 mm and 2.0 mm, and the 6 total nitrogen concentration grades are 0 g/kg, 0.04 g/kg, 0.08 g/kg, 0.12 g/kg, 0.16 g/kg and 0.2 g/kg.

In a second aspect, an embodiment of the present invention provides a device for eliminating the influence of soil particle size on soil parameters detected in discrete near-infrared bands, including:

The absorbance determination unit is used to determine the original absorbance of the soil to be detected in multiple detection bands under near-infrared spectral scanning;

A coefficient determination unit, configured to determine the particle size correction coefficient of the soil to be detected based on the original absorbance of the soil to be detected in the characteristic waveband; the characteristic waveband is based on the near-infrared spectrum scanning of soil samples with multiple soil particle sizes The original absorbance of the samples of multiple preset bands is determined;

The correction detection unit is configured to correct the original absorbance of each detection band of the soil to be detected based on the particle size correction coefficient of the soil to be detected, and determine soil parameters of the soil to be detected based on the correction result.

In a third aspect, an embodiment of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the first aspect when the processor executes the computer program The steps of the method for eliminating the influence of soil particle size on detection of soil parameters in discrete near-infrared bands.

In a fourth aspect, an embodiment of the present invention provides a non-transitory computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, implements the soil particle size versus discrete near-infrared method described in the first aspect. Steps in the elimination method of band detection soil parameter effects.

In the method for eliminating the influence of soil particle size on soil parameters detected in discrete near-infrared wavelength bands provided by the embodiment of the present invention, the particle size correction coefficient of the soil to be detected is determined according to the original absorbance of the soil to be detected in the characteristic band under near-infrared spectral scanning. The original absorbance of the detection band is corrected, and the soil parameters of the soil to be detected are determined based on the correction results, which reduces the interference of soil particle size on the discrete near-infrared band and improves the detection accuracy of soil parameters.

Description of drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description These are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained according to these drawings without creative efforts.

1 is a schematic flowchart of a method for eliminating the influence of soil particle size on discrete near-infrared band detection soil parameters provided by an embodiment of the present invention;

2 is a schematic flowchart of a method for detecting soil total nitrogen concentration provided in an embodiment of the present invention;

3 is a graph showing the relationship between a standard deviation value and a preset band provided by an embodiment of the present invention;

4 is a schematic diagram of a classification result of a soil particle size classification model provided by an embodiment of the present invention;

5 is a schematic diagram of the original absorbance of a soil total nitrogen concentration detection sample provided by an embodiment of the present invention;

6 is a schematic diagram of the corrected absorbance of a soil total nitrogen concentration detection sample provided in an embodiment of the present invention;

7 is a schematic diagram of a prediction result of soil total nitrogen concentration based on original absorbance provided by an embodiment of the present invention;

8 is a schematic diagram of a prediction result of soil total nitrogen concentration based on corrected absorbance provided by an embodiment of the present invention;

9 is a schematic diagram of a prediction result of soil total nitrogen concentration based on reference absorbance provided by an embodiment of the present invention;

10 is a schematic structural diagram of a device for eliminating the influence of soil particle size on soil parameters detected in discrete near-infrared bands according to an embodiment of the present invention;

FIG. 11 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention.

Detailed ways

In order to make the purposes, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments These are some embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

In recent years, with the needs of variable fertilization operations, smart farmland management and sustainable agricultural development, it has become more and more urgent to provide real-time and accurate detection of farmland soil parameters. The detection of soil parameters by traditional laboratory chemical methods is not only time-consuming, costly, but also polluting the environment. It is a more cost-effective choice to use spectroscopic technology for rapid and non-destructive testing of soil parameters. However, most of the spectrometers used in laboratories and soil parameter detectors based on spectroscopic technology mostly use spectrometers as components to detect soil parameters. The instrument is not only expensive but also not suitable for the harsh environment of farmland. The detection of soil parameters by a soil parameter detector based on discrete near-infrared bands has become a current research hotspot. However, this method faces serious interference caused by soil particle size.

In view of the deficiencies of the prior art, FIG. 1 is a schematic flowchart of a method for eliminating the influence of soil particle size on soil parameters detected in discrete near-infrared bands according to an embodiment of the present invention. As shown in FIG. 1 , the method includes:

Step 110, determining the original absorbance of the soil to be detected under the near-infrared spectrum scanning of multiple detection bands;

Specifically, by detecting the factors affecting the environmental quality of farmland soil, the environmental quality of farmland and its change trend can be determined. Soil testing parameters can include total nitrogen, total phosphorus, organic matter, available boron, etc. The embodiments of the present invention do not specifically limit the soil detection parameters, and the following content will be described by taking the detection of soil total nitrogen concentration as an example.

When scanning the soil to be detected, a vehicle-mounted soil total nitrogen detector based on discrete near-infrared bands can be selected. Using the near-infrared spectrum of multiple detection bands to scan the soil to be detected, the original absorbance of the soil to be detected under the scan of the near-infrared spectrum of each detection band is obtained.

Multiple detection bands of near-infrared spectral scanning can be selected according to actual measurement requirements.

For example, to detect the total nitrogen concentration of the soil to be detected, the detection bands of the near-infrared spectrum can be selected from 1070 nm, 1130 nm, 1245 nm, 1375 nm, 1550 nm, and 1680 nm.

Step 120: Determine the particle size correction coefficient of the soil to be detected based on the original absorbance of the soil to be detected in the characteristic waveband; the characteristic waveband is the original sample of the soil sample in multiple preset wavebands under the near-infrared spectral scanning based on multiple soil particle sizes. Absorbance is determined;

Specifically, when the soil to be detected is scanned by near-infrared spectroscopy, the soil particle size will interfere with the original absorbance obtained by scanning to a certain extent. The particle size correction coefficient is a coefficient for correcting the original absorbance obtained by near-infrared spectroscopy in order to eliminate the measurement deviation caused by the interference of soil particle size as much as possible.

The characteristic band is the band that can reflect the soil particle size through the corresponding raw absorbance. The characteristic band is determined based on the original absorbance of the soil samples in multiple preset bands under the near-infrared spectral scanning of the soil samples under multiple soil particle sizes. The raw absorbance obtained when the sample was scanned. The embodiment of the present invention does not specifically limit the number of characteristic bands, and preferably, two can be selected for the number of characteristic bands.

Determine the particle size correction coefficient of the soil to be tested according to the original absorbance of the characteristic wavelength bands. For example, the particle size correction coefficient of the soil to be tested can be determined according to the average value of the original absorbance of the characteristic wavelength bands of 1361 nm and 1870 nm, or the original absorbance of the characteristic wavelength bands of 1361 nm and 1870 nm. The difference between the absorbance and the original absorbance of other wavelength bands determines the particle size correction coefficient of the soil to be tested.

Step 130: Correct the original absorbance of each detection band of the soil to be detected based on the particle size correction coefficient of the soil to be detected, and determine soil parameters of the soil to be detected based on the correction result.

Specifically, according to the particle size correction coefficient of the soil to be detected, the original absorbance of each detection band of the soil to be detected is corrected to obtain a correction result, that is, the corrected absorbance of each detection band of the soil to be detected, and the soil to be detected is determined based on the correction result. soil parameters.

For example, the particle size correction coefficient P of the soil to be tested is determined according to the original absorbance of the characteristic bands of 1361 nm and 1870 nm, and the particle size correction coefficient P is used to correct the original absorbance of the bands of 1070 nm, 1130 nm, 1245 nm, 1375 nm, 1550 nm and 1680 nm respectively. The modified absorbance of the detection band was used to establish a soil total nitrogen prediction model using BP neural network, and the corrected absorbance of each detection band was used as a parameter to input the prediction model to predict the soil total nitrogen concentration.

In the method for eliminating the influence of soil particle size on soil parameters detected in discrete near-infrared wavelength bands provided by the embodiment of the present invention, the particle size correction coefficient of the soil to be detected is determined according to the original absorbance of the soil to be detected in the characteristic band under near-infrared spectral scanning. The original absorbance of the detection band is corrected, and the soil parameters of the soil to be detected are determined based on the correction results, which reduces the interference of soil particle size on the discrete near-infrared band and improves the detection accuracy of soil parameters.

Based on the above embodiment, step 120 includes:

Based on the original absorbance of the soil to be tested in the characteristic wavelength band, determine the absorbance ratio used to characterize the soil particle size of the soil to be tested;

Based on the absorbance ratio of the soil to be tested and the absorbance ratio of the reference soil particle size, the particle size correction coefficient of the soil to be tested is determined.

Specifically, the original absorbances of the characteristic wavelength bands 1361 nm and 1870 nm are A ₁₃₆₁ and A ₁₈₇₀ respectively, and the absorbance ratio R representing the soil particle size of the soil to be tested is determined, which is expressed by the formula as:

则可以确定待检测土壤的粒度修正系数P，用公式表示为：The absorbance ratio of the reference soil particle size is

Then the particle size correction coefficient P of the soil to be tested can be determined, which is expressed as:

Based on any of the above embodiments, the absorbance ratio of the reference soil particle size is determined based on the original absorbance of the soil sample under the near-infrared spectral scanning of the characteristic wavelength band of the soil sample under the reference soil particle size.

Specifically, soil with a soil particle size of 0.2 mm can be selected as the reference soil. Using multiple soil samples with a soil particle size of 0.2 mm, for each soil sample, the original absorbance of the sample was obtained under the near-infrared spectrum scanning of the characteristic band, and the absorbance ratio of the soil sample was further obtained. The average value of the absorbance ratios of multiple soil samples was taken as the absorbance ratio of the reference soil particle size.

Based on any of the above embodiments, the characteristic band is obtained based on the following method:

Perform near-infrared spectral scanning on soil samples under multiple soil particle sizes to determine the original absorbance of the soil samples in multiple preset bands;

Determine the standard deviation value of the soil sample under each soil particle size and the comprehensive standard deviation value based on the original absorbance of the sample in multiple preset bands of each soil sample;

Based on the standard deviation value of the soil sample at each soil particle size, and the integrated standard deviation value, the characteristic band is determined.

Specifically, near-infrared spectrum scanning is performed on soil samples under multiple soil particle sizes to determine the original absorbance of the soil samples in multiple preset wavelength bands. According to the definition of the near-infrared spectral region by the American Society for Testing and Materials (ASTM), the near-infrared spectrum is the region of 780nm-2526nm. In the embodiment of the present invention, the preset wavelength band is a wavelength band within 850 nm-2500 nm of the near-infrared spectrum. Each preset waveband is an equally spaced waveband, and the start and end points of each waveband are connected in sequence to achieve maximum coverage of the near-infrared spectral range.

According to the original absorbance of the sample in multiple preset bands of the soil sample, the standard deviation value of the soil sample under each soil particle size and the comprehensive standard deviation value are determined. The comprehensive standard deviation value is the standard deviation value of the soil sample as a whole.

可以通过如下公式得到：Here, the standard deviation value of the soil sample at any soil particle size j

It can be obtained by the following formula:

为土壤粒度等级为j的土壤样本在预设波段为i的近红外光谱扫描下的样本原始吸光度，

为土壤粒度等级为j的N个土壤样本在预设波段为i的近红外光谱扫描下的样本原始吸光度的平均值。In the formula, i is the serial number of the preset band, j is the soil particle size grade serial number, N is the number of soil samples under any soil particle size, W is the serial number of soil samples under any soil particle size, W==1,2,3… N,

is the original absorbance of the soil sample with the soil particle size class j under the near-infrared spectral scan of the preset band i,

is the average value of the original absorbance of the samples under the near-infrared spectral scanning of the preset band i of N soil samples with soil particle size grade j.

According to the standard deviation value of the soil sample under each soil particle size and the comprehensive standard deviation value, the characteristic band is determined. For example, the relationship curve between the standard deviation value of the soil sample under each soil particle size and the preset band, and the relationship curve between the integrated standard deviation value and the preset band can be drawn to determine the characteristic band.

Based on any of the above embodiments, based on the standard deviation value of the soil sample under each soil particle size and the comprehensive standard deviation value, the characteristic band is determined, and the following further includes:

Determine the sample absorbance ratio of the soil sample based on the sample original absorbance of the characteristic band of the soil sample;

Based on the sample absorbance ratio and soil particle size of soil samples, a soil particle size classification model was established to verify the ability of the sample absorbance ratio to characterize soil particle size.

Specifically, the sample absorbance ratio of the soil sample is determined according to the original sample absorbance of the characteristic waveband of the soil sample.

In order to verify the ability of the sample absorbance ratio to characterize soil particle size, a soil particle size classification model was established, and the sample absorbance ratio of the soil sample was used as an input variable to obtain the soil particle size classification result, which was compared with the soil particle size of the soil sample. The soil particle size classification model may be established based on an SVM (Support Vector Machine, support vector machine) model, and the embodiment of the present invention does not specifically limit the selection of the soil particle size classification model.

Based on any of the above embodiments, the characteristic wavelength bands are 1361 nm and 1870 nm.

Specifically, after using the near-infrared spectrum in the range of 850nm-2500nm to scan soil samples with different soil particle sizes, the characteristic wavelength bands are 1361nm and 1870nm after analysis using the method in the above embodiment.

Based on any of the above embodiments, the soil sample includes 4 particle size grades and 6 total nitrogen concentration gradients, 4 particle size grades are 0.2mm, 0.45mm, 0.9mm and 2.0mm, 6 total nitrogen concentration grades are 0g/kg, 0.04g/kg, 0.08g/kg, 0.12g/kg, 0.16g/kg and 0.2g/kg.

The above method will be described below through an example of soil total nitrogen concentration detection. 2 is a schematic flowchart of a method for detecting soil total nitrogen concentration provided by an embodiment of the present invention. As shown in FIG. 2 , the detailed execution steps of the soil total nitrogen concentration detection process are as follows:

In the first step, the sampled standard soil (with a total nitrogen concentration of 0 g/kg) was dried at a drying temperature of 85°C and a drying time of 24 hours. The dried soil samples were divided into 6 groups of 4 each. Soil samples were prepared using ammonia solutions made from urea, with concentrations ranging from 1 to 6. The ammonia content of grade 1 is 0g/kg, the nitrogen content of grade 2 is 0.04g/kg, the nitrogen content of grade 3 is 0.08g/kg, the nitrogen content of grade 4 is 0.12g/kg, and the nitrogen content of grade 5 is 0.16g/kg g/kg, grade 6 nitrogen content is 0.2g/kg.

When configuring the soil samples, the soil samples can also be configured by simulating the average soil moisture content (7%) of the actual farmland plough layer in summer.

After the soil configuration is completed, drying treatment is performed, and all soil samples are sieved into 10-mesh sieve (2.0mm), 20-mesh sieve (0.9mm), 40-mesh sieve (0.45mm) and 80-mesh sieve (0.2mm) ). Finally, 4 groups of soil samples with different particle sizes were obtained, each group contained 6 levels of total nitrogen concentration, and each level of total nitrogen concentration contained 4 soil samples. A total of 96 soil samples with different particle sizes and total nitrogen concentrations were obtained. Table 1 shows the data obtained after statistical analysis of soil samples.

Table 1 Statistical analysis of soil samples

其中，i为预设波段序号，波段间隔为3nm，则i＝850,853,856…,2500。m为土壤样本序号，m＝1,2,3…,96。In the second step, 96 soil samples with 4 soil particle size gradients and 6 total nitrogen concentration levels were scanned by near-infrared spectroscopy to obtain the original absorbance values of the samples in the range of 850nm-2500nm.

Wherein, i is the preset band sequence number, and the band interval is 3nm, then i=850, 853, 856..., 2500. m is the soil sample serial number, m=1,2,3...,96.

并计算所有96个土壤样本的综合标准偏差值

j是土壤粒度等级序号，j＝1为0.2mm下全部24个土壤样本，j＝2为0.45mm下全部24个土壤样本，j＝3为0.9mm下全部24个土壤样本，j＝4为2.0mm下全部24个土壤样本。The third step is to calculate the standard deviation value separately for the soil samples under 4 soil particle sizes

and calculate the combined standard deviation value for all 96 soil samples

j is the soil particle size grade number, j=1 is all 24 soil samples under 0.2mm, j=2 is all 24 soil samples under 0.45mm, j=3 is all 24 soil samples under 0.9mm, j=4 is All 24 soil samples under 2.0mm.

FIG. 3 is a graph showing the relationship between the standard deviation value and the preset band provided by the embodiment of the present invention. As shown in FIG. 3 , there are 5 curves in total, which are the standard deviations of soil samples under 4 soil particle sizes respectively. The relationship curve between the value and the preset band, as well as the relationship curve between the comprehensive standard deviation value and the preset band, the characteristic bands of soil particle size can be obtained as 1361nm and 1870nm.

The fourth step is to determine the sample absorbance ratio R ^m of the soil sample according to the original absorbance of the soil sample in the characteristic wavelength bands of 1361 nm and 1870 nm, which is expressed as:

为土壤样本在特征波段1870nm的样本原始吸光度，

为土壤样本在特征波段1361nm的样本原始吸光度。where m is the soil sample serial number,

is the original absorbance of the soil sample in the characteristic wavelength band 1870nm,

It is the original absorbance of the soil sample at the characteristic wavelength band 1361nm.

The fifth step is to establish a soil particle size classification model based on SVM (Support Vector Machine), and use the sample absorbance ratio R ^m as a single input variable to verify the accuracy of the ratio R ^m for soil particle size classification.

FIG. 4 is a schematic diagram of a classification result of a soil particle size classification model provided by an embodiment of the present invention. As shown in FIG. 4 , the predicted soil particle size (Predicted soil particle size) obtained by the soil particle size classification model is relative to the actual soil particle size (Measured soil particle size). The sample absorbance ratio ^Rm can well characterize the classification of soil particle size. Statistics on the accuracy of soil particle size classification are shown in Table 2:

Table 2 Accuracy statistics of soil particle size classification

serial number Soil particle size classification Classification accuracy 1 0.20mm 100% 2 0.45mm 83.3% 3 0.90mm 91.7% 4 2.00mm 100%

It can be seen that the accuracy of the sample absorbance ratio ^Rm determined according to the original absorbance of the soil samples in the characteristic wavelength bands of 1361nm and 1870nm for soil particle size classification is 93.8%, which meets the needs of practical applications.

The sixth step, based on the original absorbance of the characteristic band, determine the particle size correction coefficient P ^m of the soil sample, which can be expressed as:

为土壤样本在特征波段1870nm的样本原始吸光度，

为土壤样本在特征波段1361nm的样本原始吸光度，

为以0.2mm为基准土壤粒度下土壤样本在1870nm和1361nm处的吸光度比值平均值。where m is the soil sample serial number,

is the original absorbance of the soil sample in the characteristic wavelength band 1870nm,

is the original absorbance of the soil sample in the characteristic wavelength band 1361nm,

is the average value of the absorbance ratio of soil samples at 1870nm and 1361nm with 0.2mm as the benchmark soil particle size.

进行修正，获得修正后的土壤吸光度值

用公式可以表示为：The seventh step is to use the particle size correction coefficient P ^m to obtain the original absorbance value of the soil sample

Make corrections to obtain the corrected soil absorbance value

The formula can be expressed as:

In the formula, i is the preset band serial number, m is the soil sample serial number.

Figure 5 is a schematic diagram of the original absorbance of the soil total nitrogen concentration detection sample provided by the embodiment of the present invention. As shown in Figure 5, when the soil total nitrogen concentration is 0.068g/kg, a single soil sample has six discrete near-infrared bands (1070nm, 1130nm , 1245nm, 1375nm, 1550nm and 1680nm), the absorbance values of the original soil samples (Original spectrum), the soil particle size (Soil particle size) were 2.0mm, 0.9mm, 0.45mm and 0.2mm. The raw absorbance value of the sample.

Figure 6 is a schematic diagram of the corrected absorbance of the soil total nitrogen concentration detection sample provided by the embodiment of the present invention. As shown in Figure 6, when the soil total nitrogen concentration is 0.068g/kg, a single soil sample has six discrete near-infrared bands (1070nm, 1130nm) , 1245nm, 1375nm, 1550nm and 1680nm), the absorbance values of the original soil samples (Original spectrum), the soil particle size (Soil particle size) were 2.0mm, 0.9mm, 0.45mm and 0.2mm. The sample corrected absorbance value of the sample.

The eighth step, use the vehicle-mounted soil total nitrogen detector based on discrete near-infrared bands, select the near-infrared spectrum of 1070nm, 1130nm, 1245nm, 1375nm, 1550nm and 1680nm to scan the soil to be tested to obtain the original absorbance of the soil to be tested. .

According to the methods in the first to seventh steps, the original absorbance of the characteristic wavebands 1361nm and 1870nm is used to determine the particle size correction coefficient of the soil to be tested, and the original absorbance of the soil to be tested is corrected to obtain the corrected absorbance of the soil to be tested.

In addition, as a comparison, the absorbance of the soil to be tested at a soil particle size of 0.2 mm was used as the reference absorbance.

The BP neural network is used to establish the prediction model of soil total nitrogen concentration, and the original absorbance, corrected absorbance and reference absorbance of the soil to be tested are respectively input into the prediction model of soil total nitrogen concentration, and the respective prediction results of total nitrogen concentration are obtained, as shown in Table 3:

Table 3 Model accuracy statistics based on different soil absorbances

Fig. 7 is a schematic diagram of a prediction result of soil total nitrogen concentration based on original absorbance provided by an embodiment of the present invention, Fig. 8 is a schematic diagram of a predicted result of soil total nitrogen concentration based on corrected absorbance provided by an embodiment of the present invention, and Fig. 9 is provided by an embodiment of the present invention The schematic diagram of the prediction results of soil total nitrogen concentration based on reference absorbance, as shown in Figure 7, Figure 8 and Figure 9, compared with the prediction results of soil total nitrogen concentration based on reference absorbance (Figure 9), the original absorbance-based soil total nitrogen concentration The concentration prediction result (Fig. 7) has a large error, and the soil total nitrogen concentration prediction result (Fig. 8) based on the corrected absorbance is closer to the actual situation, which reduces the interference of soil particle size on discrete near-infrared bands and improves the detection accuracy of soil parameters. .

Among them, the measured values of soil total nitrogen in Figure 7, Figure 8 and Figure 9 are obtained by using a Kjeldahl nitrogen analyzer to measure the soil to be tested.

Based on any of the above embodiments, FIG. 10 is a schematic structural diagram of a device for eliminating the influence of soil particle size on soil parameters detected in discrete near-infrared bands according to an embodiment of the present invention. As shown in FIG. 10 , the device includes:

an absorbance determination unit 1010, configured to determine the original absorbance of the soil to be detected in multiple detection bands under near-infrared spectral scanning;

The coefficient determination unit 1020 is configured to determine the particle size correction coefficient of the soil to be detected based on the original absorbance of the soil to be detected in the characteristic waveband; the characteristic waveband is a plurality of presets based on the near-infrared spectrum scanning of soil samples with multiple soil particle sizes The original absorbance of the sample in the wavelength band is determined;

The correction detection unit 1030 is configured to correct the original absorbance of each detection band of the soil to be detected based on the particle size correction coefficient of the soil to be detected, and determine soil parameters of the soil to be detected based on the correction result.

Specifically, the absorbance determination unit 1010 is configured to determine the original absorbance of the soil to be detected in multiple detection bands under near-infrared spectral scanning. The coefficient determination unit 1020 is configured to determine the particle size correction coefficient of the soil to be detected based on the original absorbance of the soil to be detected in the characteristic waveband. The correction detection unit 1030 is configured to correct the original absorbance of each detection band of the soil to be detected based on the particle size correction coefficient of the soil to be detected, and determine soil parameters of the soil to be detected based on the correction result.

The device for eliminating the influence of soil particle size on soil parameters detected in discrete near-infrared wavelength bands provided by the embodiment of the present invention determines the particle size correction coefficient of the soil to be detected according to the original absorbance of the characteristic band of the soil to be detected under near-infrared spectral scanning, and determines the particle size correction coefficient of the soil to be detected. The original absorbance of the detection band is corrected, and the soil parameters of the soil to be detected are determined based on the correction results, which reduces the interference of soil particle size on the discrete near-infrared band and improves the detection accuracy of soil parameters.

Based on any of the above embodiments, the coefficient determination unit 1020 includes:

The ratio determination subunit is used to determine the absorbance ratio used to characterize the soil particle size of the soil to be detected based on the original absorbance of the soil to be detected in the characteristic wavelength band;

The coefficient determination subunit is used for determining the particle size correction coefficient of the soil to be tested based on the absorbance ratio of the soil to be tested and the absorbance ratio of the reference soil particle size.

Based on any of the above embodiments, the absorbance ratio of the reference soil particle size is determined based on the original absorbance of the soil sample under the near-infrared spectral scanning of the characteristic wavelength band of the soil sample under the reference soil particle size.

Based on any of the above embodiments, the characteristic band is obtained based on the following method:

Perform near-infrared spectral scanning on soil samples under multiple soil particle sizes to determine the original absorbance of the soil samples in multiple preset bands;

Determine the standard deviation value of the soil sample under each soil particle size and the comprehensive standard deviation value based on the original absorbance of the sample in multiple preset bands of each soil sample;

Based on the standard deviation value of the soil sample at each soil particle size, and the integrated standard deviation value, the characteristic band is determined.

Based on any of the above embodiments, based on the standard deviation value of the soil sample under each soil particle size and the comprehensive standard deviation value, the characteristic band is determined, and the following further includes:

Determine the sample absorbance ratio of the soil sample based on the original sample absorbance of the characteristic band of the soil sample;

Based on the sample absorbance ratio and soil particle size of soil samples, a soil particle size classification model was established to verify the ability of the sample absorbance ratio to characterize soil particle size.

Based on any of the above embodiments, the characteristic wavelength bands are 1361 nm and 1870 nm.

Based on any of the above embodiments, the soil sample includes 4 particle size grades and 6 total nitrogen concentration gradients, 4 particle size grades are 0.2mm, 0.45mm, 0.9mm and 2.0mm, 6 total nitrogen concentration grades are 0g/kg, 0.04g/kg, 0.08g/kg, 0.12g/kg, 0.16g/kg and 0.2g/kg.

FIG. 11 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. As shown in FIG. 11 , the electronic device may include: a processor (Processor) 1110, a communication interface (Communications Interface) 1120, a memory (Memory) 1130, and a communication bus (Communications Bus) 1140. The processor 1110 , the communication interface 1120 and the memory 1130 communicate with each other through the communication bus 1140 . The processor 1110 may invoke logic commands in the memory 1130 to perform the following methods:

Determine the original absorbance of the soil to be detected in multiple detection bands under near-infrared spectral scanning; determine the particle size correction coefficient of the soil to be detected based on the original absorbance of the soil to be detected in the characteristic band; the characteristic band is based on the soil under multiple soil particle sizes. The original absorbance of the sample in multiple preset bands under near-infrared spectral scanning is determined; based on the particle size correction coefficient of the soil to be detected, the original absorbance of each detection band of the soil to be detected is corrected, and based on the correction results, the to-be-detected is determined. Soil parameters of the soil.

In addition, the above-mentioned logic commands in the memory 1130 can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product in essence, or the part that contributes to the prior art or the part of the technical solution. The computer software product is stored in a storage medium, including Several commands are used to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, Read-Only Memory (ROM, Read-Only Memory), Random Access Memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program codes .

Embodiments of the present invention further provide a non-transitory computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, it is implemented to perform the methods provided by the foregoing embodiments, for example, including:

Determine the original absorbance of the soil to be detected in multiple detection bands under near-infrared spectral scanning; determine the particle size correction coefficient of the soil to be detected based on the original absorbance of the soil to be detected in the characteristic band; the characteristic band is based on the soil under multiple soil particle sizes. The original absorbance of the sample in multiple preset bands under near-infrared spectral scanning is determined; based on the particle size correction coefficient of the soil to be detected, the original absorbance of each detection band of the soil to be detected is corrected, and based on the correction results, the to-be-detected is determined. Soil parameters of the soil.

The device embodiments described above are only illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in One place, or it can be distributed over multiple network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution in this embodiment. Those of ordinary skill in the art can understand and implement it without creative effort.

From the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and certainly can also be implemented by hardware. Based on this understanding, the above-mentioned technical solutions can be embodied in the form of software products in essence or the parts that make contributions to the prior art, and the computer software products can be stored in computer-readable storage media, such as ROM/RAM, magnetic A disc, an optical disc, etc., includes several commands to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to perform the methods described in various embodiments or some parts of the embodiments.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, but not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand: it can still be Modifications are made to the technical solutions described in the foregoing embodiments, or some technical features thereof are equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (10)

1. A method for eliminating the influence of soil granularity on the detection of soil parameters in a discrete near-infrared band is characterized by comprising the following steps:

determining the original absorbance of the soil to be detected in a plurality of detection wave bands under near infrared spectrum scanning;

determining a granularity correction coefficient of the soil to be detected based on the original absorbance of the soil to be detected in the characteristic wave band; the characteristic wave band is determined based on the original absorbance of samples of a plurality of preset wave bands of soil samples under a plurality of soil granularity under near infrared spectrum scanning;

and correcting the original absorbance of each detection wave band of the soil to be detected based on the granularity correction coefficient of the soil to be detected, and determining the soil parameters of the soil to be detected based on the correction result.

2. The method for eliminating the influence of the soil granularity on the soil parameter detection in the discrete near-infrared band according to claim 1, wherein the determining the granularity correction coefficient of the soil to be detected based on the original absorbance of the soil to be detected in the characteristic band comprises:

determining an absorbance ratio for representing the soil granularity of the soil to be detected based on the original absorbance of the soil to be detected in the characteristic wave band;

and determining the granularity correction coefficient of the soil to be detected based on the absorbance ratio of the soil to be detected and the absorbance ratio of the granularity of the reference soil.

3. The method according to claim 2, wherein the absorbance ratio of the reference soil particle size is determined based on the original absorbance of the soil sample at the reference soil particle size in the characteristic wavelength band under the near infrared spectrum scan.

4. The method for eliminating the influence of the soil granularity on the soil parameter detection of the discrete near infrared band according to claim 1, wherein the characteristic band is obtained based on the following method:

performing near infrared spectrum scanning on the soil samples under the multiple soil granularities, and determining the original absorbance of the samples of multiple preset wave bands of the soil samples;

determining a standard deviation value and a comprehensive standard deviation value of each soil sample under each soil granularity based on the original absorbance of the sample of each soil sample at a plurality of preset wave bands;

and determining the characteristic wave band based on the standard deviation value of the soil sample under each soil granularity and the comprehensive standard deviation value.

5. The method for eliminating influence of soil granularity on soil parameter detection of discrete near infrared band according to claim 4, wherein the determining the characteristic band based on the standard deviation value of the soil sample at each soil granularity and the comprehensive standard deviation value further comprises:

determining a sample absorbance ratio of the soil sample based on the original sample absorbance of the characteristic wave band of the soil sample;

and establishing a soil granularity classification model based on the sample absorbance ratio and the soil granularity of the soil sample so as to verify the characterization capability of the sample absorbance ratio on the soil granularity.

6. The method for eliminating influence of soil particle size on soil parameter detection in a discrete near infrared band according to any one of claims 1 to 5, wherein the characteristic bands are 1361nm and 1870 nm.

7. The method of eliminating influence of soil particle size on soil parameter detection in discrete near infrared band according to any one of claims 1 to 5, wherein the soil sample comprises 4 particle size grades and 6 total nitrogen concentration gradients, wherein 4 particle size grades are 0.2mm, 0.45mm, 0.9mm and 2.0mm, and 6 total nitrogen concentration grades are 0g/kg, 0.04g/kg, 0.08g/kg, 0.12g/kg, 0.16g/kg and 0.2 g/kg.

8. The utility model provides a soil particle size is to discrete near-infrared wave band detection soil parameter influence's remove device which characterized in that includes:

the absorbance determining unit is used for determining the original absorbance of the soil to be detected in a plurality of detection wave bands under the near infrared spectrum scanning;

the coefficient determining unit is used for determining the granularity correction coefficient of the soil to be detected based on the original absorbance of the soil to be detected in the characteristic wave band; the characteristic wave band is determined based on the original absorbance of samples of a plurality of preset wave bands of soil samples under a plurality of soil granularity under near infrared spectrum scanning;

and the correction detection unit is used for correcting the original absorbance of each detection waveband of the soil to be detected based on the granularity correction coefficient of the soil to be detected and determining the soil parameters of the soil to be detected based on the correction result.

9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the computer program implements the steps of the method for eliminating the effect of soil granularity on soil parameter detection in discrete near infrared bands as claimed in any one of claims 1 to 7.

10. A non-transitory computer readable storage medium, having stored thereon a computer program, wherein the computer program, when being executed by a processor, implements the steps of the method for eliminating the effect of soil particle size on discrete near-infrared band detection soil parameters as claimed in any one of claims 1 to 7.

Images