Browsing by Author "Dang, Tommy (TTU)"
Now showing 1 - 3 of 3
- Results Per Page
- Sort Options
Item Comparative analysis and visualization of soil profiles at the meter spatial scale utilizing novel matrix and volume rendering techniques(2023) Gonzalez, Jake (TTU); Siebecker, Matthew (TTU); Pham, Vung; Jordan, Cynthia (TTU); Weindorf, David C. (TTU); Dang, Tommy (TTU)This research introduces a soil characterization technique involving four data visualization tools to help researchers and stakeholders interpret high dimensional soil data at the field scale. This technique involves visualizing a reduced dimensionality representation of elemental concentration and color data gathered via portable X-ray fluorescence (pXRF) spectrometer and NixPro color proximal sensors, respectively. Soil cores were collected from sites located in Lubbock and Lamb Counties, West Texas, USA. Thirteen core samples were collected from these sites in a star pattern with readings from proximal sensors at depths ranging between 0 and 100 cm at 10 cm intervals. The dimensionality reduction techniques utilize four visualization tools to represent soil composition data through multiple user-adjustable variables (i.e., mg kg−1 elemental concentrations and soil profiles), offering more insight and control compared to a single-variable approach. Through these tools and techniques, qualitative and quantitative conclusions regarding soil characteristics (e.g., elemental concentration variation, delineation of soil horizons, changes in soil color) can be formulated from the data and used in various applications. Areas where these novel software tools can be utilized potentially include rapid contaminant mapping in soils, characterization of diagnostic soil horizons (e.g., calcic, spodic, gypsic, etc.), micronutrient distribution at a field scale for precision agricultural purposes, and pedometrics.Item iDVS: interactive 2D and 3D visualizations of proximal sensor data for rapid characterization of soil profiles(2023) Pham, Vung; Jordan, Cynthia M. (TTU); Siebecker, Matthew G. (TTU); Weindorf, David C.; Dang, Tommy (TTU)Knowledge of the soil’s physical and chemical properties in field-scale geographical areas is crucial for farmers and policymakers for agronomic productivity and environmental quality assessment. Proximal sensors can successfully model soil properties for these purposes and offer a way to rapidly acquire data from soil profiles. However, existing data analysis approaches are ill-suited to explore this type of multivariate proximal sensor data over large land areas and in a sizeable three-dimensional volume. Therefore, this work proposes a multifaceted approach with seamless integration of a star pattern for soil sample collection, data acquisition using proximal sensor devices, and an interactive data visualization solution for processing, analyzing, and reporting analysis results. This solution is the result of an interdis- ciplinary project in which data visualizers worked closely with soil scientists and agronomists to develop a tool called iDVS for rapid characterizations of soil profiles over larger geographical areas using proximal sensor technologies.Item Soil profile analysis using interactive visualizations, machine learning, and deep learning(2021) Pham, Vung; Weindorf, David C.; Dang, Tommy (TTU)Soil is an essential element of life, and soil properties are crucial in analyzing soil health. Recent developments of proximal sensor technologies, such as portable X-ray fluorescence (pXRF) spectroscopy or visible and near-infrared (Vis–NIR) spectroscopy, offer rapid and non-destructive alternatives for quantifying data from soil profiles. While the data collection time using these technologies decreases significantly, the subsequent analysis remains time-consuming, and current analysis solutions only provide basic visualizations. Furthermore, the use of collected data from proximal sensors to predict high-level soil properties has garnered worldwide attention in the past decade, owing to its convenience. Therefore, this paper discusses the objectives for software solutions in this area, consolidated from interviewing 102 stakeholders. Following these requirements, data visualizers work closely with soil scientists to propose a set of interactive visualizations for analyzing soil profiles using pXRF data. These interactive visualizations receive positive feedback from the domain experts. This project also explores various machine learning and deep learning approaches to predict soil properties from spectral data. This work then proposes a deep learning model called RDNet that achieves state-of-the-art results in predicting pHH2O and pHKCl from Vis–NIR spectra acquired from a set of globally distributed soil samples.