CN-121995036-A - Soil hardness prediction method, apparatus, device, storage medium, and program product

CN121995036ACN 121995036 ACN121995036 ACN 121995036ACN-121995036-A

Abstract

The application discloses a soil hardness prediction method, a device, equipment, a storage medium and a program product, and relates to the technical field of image data processing, wherein the soil hardness prediction method comprises the following steps: by determining the soil attribute data, predicting the soil humidity time sequence data by utilizing the attribute data and classifying the soil, the soil penetration resistance is finally determined, the comprehensive calculation of the soil hardness is realized, the change trend in a period of time in the future can be reflected, and the timeliness and the refinement degree of drawing are obviously improved. Meanwhile, widely available environmental data (such as satellite remote sensing, digital elevation model and weather re-analysis products) are used as input, and the method can be rapidly applied to areas lacking in measured data, so that the data acquisition cost and time are greatly reduced, and the method has the capability of popularization and application in large-scale areas (such as provincial and river basins).

Inventors

CHEN WEITAO
Wang Ruidie
ZHU HONGWEI

Assignees

中国地质大学（武汉）

Dates

Publication Date: 20260508
Application Date: 20260409

Claims (10)

1. A method for predicting soil hardness, the method comprising: Acquiring environmental data of an area to be predicted, and determining soil attribute data of the area to be predicted based on the environmental data; predicting the soil humidity of the area to be predicted according to the environmental data and the soil attribute data to obtain soil humidity time sequence data of the area to be predicted; performing soil classification on the area to be predicted based on the soil attribute data to obtain soil classification data of the area to be predicted; And determining the soil penetration resistance of the area to be predicted based on the soil attribute data, the soil humidity time sequence data and the soil classification data, wherein the soil penetration resistance is used for measuring the soil hardness.
2. The soil hardness prediction method according to claim 1, wherein the step of determining soil property data of the region to be predicted based on the environmental data comprises: Determining an environment covariate of the area to be predicted based on the environment data; and inputting the environment covariates into a pre-trained random forest regression model for prediction to obtain soil attribute data of the region to be predicted.
3. The method for predicting soil hardness according to claim 2, wherein before the step of inputting the environmental covariates into a pre-trained random forest regression model for prediction to obtain the soil property data of the area to be predicted, the method further comprises: Acquiring training soil attribute data and training environment covariate data, wherein the spatial scale of the training soil attribute data is a first scale, the spatial scale of the training environment covariate data is a second scale, and the first scale is larger than the second scale; performing downsampling processing based on the training environment covariate data to obtain training environment covariate data corresponding to the first scale; And training a random forest regression model based on the training soil attribute data and the training environment covariate data corresponding to the first scale to obtain a pre-trained random forest regression model.
4. A soil hardness prediction method according to claim 3, wherein the step of inputting the environmental covariates into a pre-trained random forest regression model for prediction to obtain soil property data of the region to be predicted comprises: performing downsampling treatment on the environment covariates to obtain environment covariates corresponding to the first scale; Inputting the environment covariates and the environment covariates corresponding to the first scale to the pre-trained random forest regression model for prediction to obtain soil attribute data of the first scale and soil attribute data of the second scale; and carrying out residual correction on the soil attribute data of the second scale based on the soil environment data of the first scale to obtain the soil attribute data of the area to be predicted.
5. The method for predicting soil hardness according to claim 1, wherein the step of predicting soil moisture of the area to be predicted based on the environmental data and the soil attribute data to obtain the soil moisture time series data of the area to be predicted further comprises: Acquiring training environment data, wherein the training environment data at least comprises optical remote sensing data, first soil humidity data and second soil humidity data; Performing deviation correction on the first soil humidity data based on the second soil humidity data to obtain training soil humidity data; performing image fusion based on the optical remote sensing data to obtain a fusion spectrum index; Constructing a short-term prediction model according to the fusion spectrum index, the predicted time sequence variable data and the training soil humidity data to obtain a soil humidity prediction model; correspondingly, the step of predicting the soil humidity of the area to be predicted according to the environmental data and the soil attribute data to obtain the soil humidity time sequence data of the area to be predicted comprises the following steps: And inputting the environmental data and the soil attribute data into the soil humidity prediction model to predict the soil humidity, so as to obtain the soil humidity time sequence data of the region to be predicted.
6. The method of predicting soil hardness according to claim 1, wherein before the step of classifying the soil of the area to be predicted based on the soil attribute data to obtain soil classification data of the area to be predicted, further comprising: Acquiring a first training database and a second training database, and determining training soil attribute data shared by the first training database and the second training data; performing data cleaning on the training soil attribute data based on the structured classification rule to obtain cleaned training soil attribute data; performing classification model training based on the cleaned training soil attribute data to obtain a soil classification model; correspondingly, the step of classifying the soil of the area to be predicted based on the soil attribute data to obtain soil classification data of the area to be predicted comprises the following steps: And inputting the soil attribute data into the soil classification model to classify the soil, so as to obtain the soil classification data of the area to be predicted.
7. A soil hardness prediction apparatus, characterized in that the soil hardness prediction apparatus comprises: the soil attribute determining module is used for acquiring the environmental data of the area to be predicted and determining the soil attribute data of the area to be predicted based on the environmental data; the soil humidity prediction module is used for predicting the soil humidity of the area to be predicted according to the environmental data and the soil attribute data to obtain the soil humidity time sequence data of the area to be predicted; The soil classification model is used for classifying the soil of the area to be predicted based on the soil attribute data to obtain soil classification data of the area to be predicted; And the soil hardness determining module is used for determining the soil penetration resistance of the area to be predicted based on the soil attribute data, the soil humidity time sequence data and the soil classification data, wherein the soil penetration resistance is used for measuring the soil hardness.
8. A soil hardness prediction apparatus comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the computer program being configured to implement the steps of the soil hardness prediction method as claimed in any one of claims 1 to 6.
9. A storage medium, characterized in that the storage medium is a computer-readable storage medium, on which a computer program is stored, which computer program, when being executed by a processor, carries out the steps of the soil hardness prediction method according to any one of claims 1 to 6.
10. A computer program product, characterized in that it comprises a computer program which, when executed by a processor, implements the steps of the soil hardness prediction method according to any one of claims 1 to 6.

Description

Soil hardness prediction method, apparatus, device, storage medium, and program product Technical Field The present application relates to the field of image data processing technology, and in particular, to a soil hardness prediction method, apparatus, device, storage medium, and program product. Background Soil hardness is an important physical property that describes the ability of a soil to deform or break against external forces, also known as soil compaction or soil compaction, as measured by soil penetration resistance (Soil Penetration Resistance, SPR). The traditional soil hardness measurement mainly relies on an in-situ sounding technology based on a cone penetrometer, and the probe with a sensor is penetrated into a specific soil layer depth of a soil body, so that the soil hardness is represented according to penetration resistance encountered when the probe is pressed into the soil at a uniform speed. The space coverage of the mode is insufficient, and the continuous space-time characterization requirement of the soil hardness of a large-scale area (such as a provincial administrative district or a river basin) is difficult to meet. Disclosure of Invention The application mainly aims to provide a soil hardness prediction method, a device, equipment, a storage medium and a program product, and aims to solve the technical problem that the traditional soil hardness measurement is difficult to meet the continuous space-time characterization requirement of the soil hardness of a large-scale area. In order to achieve the above object, the present application provides a soil hardness prediction method, comprising: Acquiring environmental data of an area to be predicted, and determining soil attribute data of the area to be predicted based on the environmental data; predicting the soil humidity of the area to be predicted according to the environmental data and the soil attribute data to obtain soil humidity time sequence data of the area to be predicted; performing soil classification on the area to be predicted based on the soil attribute data to obtain soil classification data of the area to be predicted; And determining the soil penetration resistance of the area to be predicted based on the soil attribute data, the soil humidity time sequence data and the soil classification data, wherein the soil penetration resistance is used for measuring the soil hardness. In an embodiment, the step of determining soil attribute data of the area to be predicted based on the environmental data includes: Determining an environment covariate of the area to be predicted based on the environment data; and inputting the environment covariates into a pre-trained random forest regression model for prediction to obtain soil attribute data of the region to be predicted. In an embodiment, before the step of inputting the environmental covariates into a pre-trained random forest regression model to predict, the step of obtaining soil attribute data of the area to be predicted further includes: Acquiring training soil attribute data and training environment covariate data, wherein the spatial scale of the training soil attribute data is a first scale, the spatial scale of the training environment covariate data is a second scale, and the first scale is larger than the second scale; performing downsampling processing based on the training environment covariate data to obtain training environment covariate data corresponding to the first scale; And training a random forest regression model based on the training soil attribute data and the training environment covariate data corresponding to the first scale to obtain a pre-trained random forest regression model. In an embodiment, the step of inputting the environmental covariates into a pre-trained random forest regression model to predict, and obtaining soil attribute data of the region to be predicted includes: performing downsampling treatment on the environment covariates to obtain environment covariates corresponding to the first scale; Inputting the environment covariates and the environment covariates corresponding to the first scale to the pre-trained random forest regression model for prediction to obtain soil attribute data of the first scale and soil attribute data of the second scale; and carrying out residual correction on the soil attribute data of the second scale based on the soil environment data of the first scale to obtain the soil attribute data of the area to be predicted. In an embodiment, before the step of predicting the soil humidity of the area to be predicted according to the environmental data and the soil attribute data to obtain the soil humidity time series data of the area to be predicted, the method further includes: Acquiring training environment data, wherein the training environment data at least comprises optical remote sensing data, first soil humidity data and second soil humidity data; Performing deviation correction on the first soil h