Urban green land soil moisture content detection method

Abstract

The invention discloses an urban green land soil moisture content detection method. The method comprises steps of preprocessing yperspectral data and panchromatic images; setting a sample plot, and measuring the moisture content of the sample; obtaining a vegetation area and a non-vegetation area through image segmentation and contour recognition, and selecting a bare soil sample of the non-vegetation area to obtain an end member spectrum curve of bare soil; extracting an exposed soil distribution area; utilizing the thermal infrared data to respectively establish inversion models for the plant coverage area and the bare soil area, and calculating constant values in the inversion models and inversion area soil moisture content in combination with the measured values. According to the method, image segmentation, contour recognition and spectral angle classification are combined, a vegetation coverage area and an exposed soil area are distinguished, two soil moisture content models are established, the soil moisture content of the urban greenbelt can be indirectly and directly detected, the adopted method is convenient and practical, large-area soil moisture content detection can beachieved, compared with most soil moisture content determination methods, the method is faster, more accurate and more comprehensive.

Description

technical field [0001] The invention relates to the field of measuring and calculating soil moisture content, in particular to a method for detecting soil moisture content in urban green spaces. Background technique [0002] Soil moisture content is an important soil physical parameter, which is an indispensable basic parameter in the research fields of ecosystem water cycle, plant growth, soil carrying capacity and so on. First, soil moisture affects the water and energy exchange between the surface and the atmosphere. In addition, water is the basic condition for the growth and development of vegetation, and soil moisture content is the guarantee for the healthy growth of vegetation, and it is also an important parameter in irrigation management and urban green space management forecast. Therefore, soil moisture content is an important indicator for evaluating land resources and a very critical parameter in precision forestry. The monitoring of soil moisture content is of...

AI Technical Summary

Problems solved by technology

However, it is difficult to distinguish soil and other surface features in multispectral images, and often produces mixed pixels, which reduces the inversion accuracy

Most of them are measured by ground object spectrometer, the measurement range is limited, and it is difficult to quickly measure the soil moisture content in a large area.

[0004] Chinese patent document CN 104697937A records a soil attribute hyperspectral identification technology method, using remote sensing data to obtain soil hyperspectral images at different times, after preprocessing, the exposed soil is obtained through supervised classification, and the surface reflectance of the exposed soil is extracted According to the surface reflectance of the bare soil, the surface reflectance inversion model of the bare soil is established, and the soil erodibility data is measured indoors and the soil K value is calculated; according to the soil K value and the surface reflectance inversion model Spectral data, establishing a hyperspectral model of soil properties that affect erodibility K, and solving the problem that hyperspectral remote sensing technology cannot be used to measure soil erodibility

The above two methods can only measure the moisture content of bare soil. For areas covered by vegetation, it is impossible to directly measure the moisture content of soil under vegetation, and there is no good method to compare the vegetation and non-vegetation areas in hyperspectral images. The division of regions has brought errors to the calculation of soil moisture content

Images