Soil classification method through combination of multi-layer perceptron neural networks with spectral characteristic parameters

A technology of spectral feature parameters and neural network model, applied in biological neural network model, instrument, character and pattern recognition, etc., can solve the problem of unable to retain the physical meaning of spectral features, to overcome the lack of clear physical meaning, fast and accurate Effects of classification, improved accuracy and speed

Inactive Publication Date: 2017-05-10
NORTHEAST AGRICULTURAL UNIVERSITY
View PDF2 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the problem that the existing method of soil classification using soil spectral reflectance cannot retain the original physical meaning of spectral features, and to provide a soil classification using a multi-layer perceptron neural network model combined with spectral characteristic parameters method

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Soil classification method through combination of multi-layer perceptron neural networks with spectral characteristic parameters
  • Soil classification method through combination of multi-layer perceptron neural networks with spectral characteristic parameters
  • Soil classification method through combination of multi-layer perceptron neural networks with spectral characteristic parameters

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0016] Specific implementation mode one: the following combination figure 1 Illustrate this embodiment, the soil classification method that utilizes multi-layer perceptron neural network model described in this embodiment in conjunction with spectral characteristic parameter, it comprises the following steps:

[0017] Step 1: Collect n soil samples, where n is an integer greater than or equal to 3; use a spectrometer to test the reflection spectra of the n soil samples respectively, and for each soil sample: collect 10 spectral curves, and perform arithmetic on the 10 spectral curves Averaging, to obtain baseline reflectance spectral data for each soil sample;

[0018] Step 2: The reference reflectance spectrum data of each soil sample is taken as an interval of 10nm, and the Gaussian model is used for spectral resampling;

[0019] Step 3: Perform envelope removal on the spectral resampling data, and obtain envelope removal data that highlights the absorption and reflection c...

specific Embodiment

[0032] Step 1: Collect 0-20cm plow layer soil samples in 13 cities and counties including Bei'an City and Baiquan County, a total of 138, as shown in Table 1. Grind the soil samples indoors, air-dry them, and pass through a 2mm sieve. Samples were tested for reflectance spectroscopy.

[0033] Table 1

[0034] Soil black soil chernozem windy sandy soil total Sample size 35 74 29 138

[0035] use 3 Portable spectrometers perform spectral tests on soil samples in a darkroom with controlled lighting conditions. The soil samples were placed in sample containers with a diameter of 12 cm and a depth of 1.8 cm, and the surface of the soil samples was scraped flat with a ruler. The light source is a halogen lamp with a power of 1000W, 100cm away from the surface of the soil sample, and the zenith angle is 30°, which provides almost parallel light to the soil sample, which is used to reduce the influence of shadow caused by soil roughness. The sensor ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a soil classification method through combination of multi-layer perceptron neural networks with spectral characteristic parameters and belongs to the technical field of soil classification. The method aims at solving the problem that the original physical significance of spectral characteristics cannot be reserved through utilization of the existing soil classification method which employs soil spectral reflectivity. The method comprises the steps of collecting soil samples and obtaining reference reflection spectral data of each soil sample; carrying out spectral resampling on the reference reflection spectral data of each soil sample by taking 10nm as an interval through utilization of a Gaussian model; carrying out envelope line removal on the spectral resampling data, thereby obtaining the envelope line removed data which highlights the absorption and reflection characteristics of reflection spectral curves; extracting m spectral characteristic parameters from the envelope line removed data; carrying out standard processing on the extracted m spectral characteristic parameters, thereby obtaining m soil classification indexes; and classifying the soil samples according to the soil classification indexes through utilization of the multi-layer perceptron neural networks. The method is used for soil classification.

Description

technical field [0001] The invention relates to a soil classification method using a multilayer perceptron neural network model combined with spectral characteristic parameters, and belongs to the technical field of soil classification. Background technique [0002] Soil classification using soil spectral reflectance characteristics can provide technical support for accelerating fine soil mapping. At present, in the methods of soil classification using soil spectral reflectance at home and abroad, principal component analysis is often used to process spectral data, extracting principal components as the input of classification models, and using K-means, support vector machines and other methods to establish classification models. However, the variables obtained by the principal component analysis method have no clear physical meaning, and the results of different studies are not comparable. [0003] Multi-layer perceptron neural networks (MLPneural networks), as a powerful ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/62G06N3/02
CPCG06N3/02G06F18/24
Inventor 刘焕军张小康张新乐王翔窦欣秦乐乐杨皓轩
Owner NORTHEAST AGRICULTURAL UNIVERSITY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap