{"id":{"repo_id":"texas-state","oai_identifier":"oai:digital.library.txst.edu:10877/18683"},"canonical_url":"https://search.dev.ndltd.org/etd/texas-state/oai:digital.library.txst.edu:10877/18683","repository":{"repo_id":"texas-state","name":"Texas State University","base_url":"https://digital.library.txst.edu/server/oai/request"},"display":{"title":"Development of a Machine Learning Algorithm for the Estimation of Soil Organic Matter from the Integration of UAV and In-Ground Soil Sensor","abstract":"Biotic wastes and living soil organisms comprise an organic substance of soil known as soil organic matter (SOM). Many agricultural soils are severely depleted in this essential nutrient because of synthetic fertilizers and pesticides over the years. Soil fertility and production can only be better understood if the amount of organic matter in the soil is measured. Combining sophisticated technology with a traditional way of assessing SOM (loss on ignition) can help estimate the SOM's value. This proposed work is based on the pretreatments of a UAV multispectral data under ideal indexes and a collection of field measurements employing in-ground sensors inside the ML (Machine Learning) approach should produce a good correlation with the laboratory LOI (Loss on Ignition) tested samples. Two methods were used to gather the data: aerial images collected by UAVs displayed the vegetative index NDVI (Normalized Difference Vegetation Index) while sensors on the ground recorded variables such as potassium, phosphorous, nitrogen, pH, humidity, and temperature. A multispectral image postprocessing tool using Pix4Dfields merged all collected data from sensors with one hundred samples. Sample data were correlated with laboratory results to construct a machine learning algorithm. Compared to the previous method, the newly created process saves considerable time when estimating SOM. A sample of algorithms was tried in this study; however, elastic net, ridge, and linear regression were the most effective. The RMSE (Root Mean Square Value) of 0.13, 0.12, 0.13 and the R2 coefficient of determination of 0.13, 0.16, 0.12 showed that the suggested model fits well and accounts for 14% of the variations in the accurate SOM data. The experiment found that a combination of sensors, an Unmanned Ariel Vehicle, and machine learning best determine SOM. A small sample size meant that the model could not forecast values accurately. Future steps involve expanding the number of samples and using soil sensors to automate data collecting.","abstract_html":"Biotic wastes and living soil organisms comprise an organic substance of soil known as soil organic matter (SOM). Many agricultural soils are severely depleted in this essential nutrient because of synthetic fertilizers and pesticides over the years. Soil fertility and production can only be better understood if the amount of organic matter in the soil is measured. Combining sophisticated technology with a traditional way of assessing SOM (loss on ignition) can help estimate the SOM's value. This proposed work is based on the pretreatments of a UAV multispectral data under ideal indexes and a collection of field measurements employing in-ground sensors inside the ML (Machine Learning) approach should produce a good correlation with the laboratory LOI (Loss on Ignition) tested samples. Two methods were used to gather the data: aerial images collected by UAVs displayed the vegetative index NDVI (Normalized Difference Vegetation Index) while sensors on the ground recorded variables such as potassium, phosphorous, nitrogen, pH, humidity, and temperature. A multispectral image postprocessing tool using Pix4Dfields merged all collected data from sensors with one hundred samples. Sample data were correlated with laboratory results to construct a machine learning algorithm. Compared to the previous method, the newly created process saves considerable time when estimating SOM. A sample of algorithms was tried in this study; however, elastic net, ridge, and linear regression were the most effective. The RMSE (Root Mean Square Value) of 0.13, 0.12, 0.13 and the R2 coefficient of determination of 0.13, 0.16, 0.12 showed that the suggested model fits well and accounts for 14% of the variations in the accurate SOM data. The experiment found that a combination of sensors, an Unmanned Ariel Vehicle, and machine learning best determine SOM. A small sample size meant that the model could not forecast values accurately. Future steps involve expanding the number of samples and using soil sensors to automate data collecting.","abstract_has_math":false,"creators":["Basutkar, Rishab"],"institution":"Texas State University","degree_name":"Master of Science","degree_level":"Masters","degree_discipline":"Engineering Management","degree_department":null,"school":null,"contributors":[],"advisors":["Khaleghian, Meysam"],"committee_chairs":[],"committee_members":["Emami, Anahita","Mix, Kenneth D."],"year":2022,"date_issued":"2022-05","date_published":"2022-05","updated_at":"2026-07-27T21:22:49Z","subjects":["machine learning","remote sensor","UAV","soil organic matter"],"languages":["en"],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/10877/18683","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Khaleghian, Meysam"]},{"key":"dc:contributor.committeemember","label":"Committee Member","values":["Emami, Anahita","Mix, Kenneth D."]},{"key":"dc:creator","label":"Author","values":["Basutkar, Rishab"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2024-05-13T18:44:11Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2024-05-13T18:44:11Z"]},{"key":"dc:date.issued","label":"Date","values":["2022-05"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Engineering Management"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Masters"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["Texas State University"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["machine learning","remote sensor","UAV","soil organic matter"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/10877/18683"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Biotic wastes and living soil organisms comprise an organic substance of soil known as soil organic matter (SOM). Many agricultural soils are severely depleted in this essential nutrient because of synthetic fertilizers and pesticides over the years. Soil fertility and production can only be better understood if the amount of organic matter in the soil is measured. Combining sophisticated technology with a traditional way of assessing SOM (loss on ignition) can help estimate the SOM's value. This proposed work is based on the pretreatments of a UAV multispectral data under ideal indexes and a collection of field measurements employing in-ground sensors inside the ML (Machine Learning) approach should produce a good correlation with the laboratory LOI (Loss on Ignition) tested samples. Two methods were used to gather the data: aerial images collected by UAVs displayed the vegetative index NDVI (Normalized Difference Vegetation Index) while sensors on the ground recorded variables such as potassium, phosphorous, nitrogen, pH, humidity, and temperature. A multispectral image postprocessing tool using Pix4Dfields merged all collected data from sensors with one hundred samples. Sample data were correlated with laboratory results to construct a machine learning algorithm. Compared to the previous method, the newly created process saves considerable time when estimating SOM. A sample of algorithms was tried in this study; however, elastic net, ridge, and linear regression were the most effective. The RMSE (Root Mean Square Value) of 0.13, 0.12, 0.13 and the R2 coefficient of determination of 0.13, 0.16, 0.12 showed that the suggested model fits well and accounts for 14% of the variations in the accurate SOM data. The experiment found that a combination of sensors, an Unmanned Ariel Vehicle, and machine learning best determine SOM. A small sample size meant that the model could not forecast values accurately. Future steps involve expanding the number of samples and using soil sensors to automate data collecting."]},{"key":"dc:format","label":"Dc Format","values":["Text"]},{"key":"dc:format.medium","label":"Dc Format Medium","values":["1 file (.pdf)"]},{"key":"dc:title","label":"Title","values":["Development of a Machine Learning Algorithm for the Estimation of Soil Organic Matter from the Integration of UAV and In-Ground Soil Sensor"]}]}],"canonical_facts":{"dc:contributor.advisor":["Khaleghian, Meysam"],"dc:contributor.committeemember":["Emami, Anahita","Mix, Kenneth D."],"dc:creator":["Basutkar, Rishab"],"dc:date.accessioned":["2024-05-13T18:44:11Z"],"dc:date.available":["2024-05-13T18:44:11Z"],"dc:date.issued":["2022-05"],"dc:description.abstract":["Biotic wastes and living soil organisms comprise an organic substance of soil known as soil organic matter (SOM). Many agricultural soils are severely depleted in this essential nutrient because of synthetic fertilizers and pesticides over the years. Soil fertility and production can only be better understood if the amount of organic matter in the soil is measured. Combining sophisticated technology with a traditional way of assessing SOM (loss on ignition) can help estimate the SOM's value. This proposed work is based on the pretreatments of a UAV multispectral data under ideal indexes and a collection of field measurements employing in-ground sensors inside the ML (Machine Learning) approach should produce a good correlation with the laboratory LOI (Loss on Ignition) tested samples. Two methods were used to gather the data: aerial images collected by UAVs displayed the vegetative index NDVI (Normalized Difference Vegetation Index) while sensors on the ground recorded variables such as potassium, phosphorous, nitrogen, pH, humidity, and temperature. A multispectral image postprocessing tool using Pix4Dfields merged all collected data from sensors with one hundred samples. Sample data were correlated with laboratory results to construct a machine learning algorithm. Compared to the previous method, the newly created process saves considerable time when estimating SOM. A sample of algorithms was tried in this study; however, elastic net, ridge, and linear regression were the most effective. The RMSE (Root Mean Square Value) of 0.13, 0.12, 0.13 and the R2 coefficient of determination of 0.13, 0.16, 0.12 showed that the suggested model fits well and accounts for 14% of the variations in the accurate SOM data. The experiment found that a combination of sensors, an Unmanned Ariel Vehicle, and machine learning best determine SOM. A small sample size meant that the model could not forecast values accurately. Future steps involve expanding the number of samples and using soil sensors to automate data collecting."],"dc:format":["Text"],"dc:format.medium":["1 file (.pdf)"],"dc:identifier.uri":["https://hdl.handle.net/10877/18683"],"dc:language.iso":["en"],"dc:subject":["machine learning","remote sensor","UAV","soil organic matter"],"dc:title":["Development of a Machine Learning Algorithm for the Estimation of Soil Organic Matter from the Integration of UAV and In-Ground Soil Sensor"],"dc:type":["Thesis"],"thesis:degree_discipline":["Engineering Management"],"thesis:degree_level":["Masters"],"thesis:degree_name":["Master of Science"],"thesis:institution_name":["Texas State University"]},"updated_at":"2026-07-27T21:22:49Z"}