	US 11,676,375 B2
	System and process for integrative computational soil mapping
Bryan Andre Fuentes Ponce, Fayetteville, AR (US); Phillip Ray Owens, Fayetteville, AR (US); and Minerva Justine Dorantes, Fayetteville, AR (US)
Assigned to BOARD OF TRUSTEES OF THE UNIVERSITY OF ARKANSAS, Little Rock, AR (US); and THE UNITED STATES OF AMERICA, AS REPRESENTED BY THE SECRETARY OF AGRICULTURE, Washington, DC (US)
Filed by Board of Trustees of the University of Arkansas, Little Rock, AR (US); and The United States of America, as represented by The Secretary of Agriculture, Washington, DC (US)
Filed on Aug. 30, 2021, as Appl. No. 17/461,281.
Claims priority of provisional application 63/072,613, filed on Aug. 31, 2020.
Prior Publication US 2022/0067336 A1, Mar. 3, 2022
Int. Cl. G06K 9/00 (2022.01); G06V 20/13 (2022.01); G06F 18/23 (2023.01); G06F 18/213 (2023.01)

CPC G06V 20/13 (2022.01) [G06F 18/213 (2023.01); G06F 18/23 (2023.01)]

18 Claims

1. A system for integrative computational soil mapping, the system comprising:

a computer having a processor and a memory; and

a software module stored in the memory, comprising executable instructions that when executed by the processor cause the processor to:

receive electronic data representing soil information for an area of interest;

automatically or semi-automatically generate environmental layers using spatial analysis related to a cluster category;

automatically or semi-automatically perform selection of environmental layers;

automatically or semi-automatically perform data dimensionality reduction for environmental layers in each cluster category;

automatically or semi-automatically perform cluster analysis for each cluster category;

automatically or semi-automatically generate generic soil-landscape classes from spatial interactions of clustering groups across the cluster categories;

assign at least one soil property measurement to each generic soil-landscape class;

generate a digital soil map of the soil property measurement for the area of interest;

output the digital soil map to an electronic display for a user;

group the environmental layers according to the cluster category;

select the environmental layers that capture a predetermined amount of variability in environmental conditions of the area of interest;

automatically or semi-automatically reduce the data dimensionality of environmental layers using the Self Organizing Map technique;

automatically or semi-automatically recognize patterns of interaction between the selected environmental layers using cluster analysis;

automatically determine an optimum number of the clustering groups in each cluster category;

automatically predict and map a statistical distribution of the selected environmental layers for each clustering group using locally weighted polynomial regression;

automatically generate a group-likelihood map that represents a likelihood of occurrence for each clustering group across the area of interest;

automatically generate a generic soil-landscape class map using the generic soil-landscape class for each clustering group;

automatically generate a single class-likelihood map representing the likelihood of occurrence for the generic soil-landscape class across the area of interest;

assign a soil sample location on the class-likelihood map for each generic soil-landscape class where an utmost likelihood of occurrence for the generic soil-landscape class across the area of interest exists;

assign the soil property measurement for each soil sample location on the class-likelihood map to each generic soil-landscape class;

automatically perform spatial modeling of the soil property measurements;

generate the digital soil map of the modeled soil property measurements; and

output the digital soil map to the electronic display for the user.