The Food and Agriculture Organization of the United Nations (FAO) is building a land cover monitoring system in Lesotho in support of ReNOKA (&lsquo:we are a river&rsquo:), the national program for integrated catchment management led by the Government of Lesotho. The aim of the system is to deliver land cover products at a national level on an annual basis that can be used for global reporting of official land cover statistics and to inform appropriate land restoration policies. This paper presents an innovative methodology that has allowed the production of five standardized annual land cover maps (2017&ndash:2021) using only a single in situ dataset gathered in the field for the reference year, 2021. A total of 10 land cover classes are represented in the maps, including specific features, such as gullies, which are under close monitoring. The mapping approach developed includes the following: (i) the automatic generation of training and validation datasets for each reporting year from a single in situ dataset: (ii) the use of a Random Forest Classifier combined with postprocessing and harmonization steps to produce the five standardized annual land cover maps: (iii) the construction of confusion matrixes to assess the classification accuracy of the estimates and their stability over time to ensure estimates&rsquo: consistency. Results show that the error-adjusted overall accuracy of the five maps ranges from 87% (2021) to 83% (2017). The aim of this work is to demonstrate a suitable solution for operational land cover mapping that can cope with the scarcity of in situ data, which is a common challenge in almost every developing country.

The Food and Agriculture Organization of the United Nations (FAO) is building a land cover monitoring system in Lesotho in support of ReNOKA (‘we are a river’), the national program for integrated catchment management led by the Government of Lesotho. The aim of the system is to deliver land cover products at a national level on an annual basis that can be used for global reporting of official land cover statistics and to inform appropriate land restoration policies. This paper presents an innovative methodology that has allowed the production of five standardized annual land cover maps (2017–2021) using only a single in situ dataset gathered in the field for the reference year, 2021. A total of 10 land cover classes are represented in the maps, including specific features, such as gullies, which are under close monitoring. The mapping approach developed includes the following: (i) the automatic generation of training and validation datasets for each reporting year from a single in situ dataset; (ii) the use of a Random Forest Classifier combined with postprocessing and harmonization steps to produce the five standardized annual land cover maps; (iii) the construction of confusion matrixes to assess the classification accuracy of the estimates and their stability over time to ensure estimates’ consistency. Results show that the error-adjusted overall accuracy of the five maps ranges from 87% (2021) to 83% (2017). The aim of this work is to demonstrate a suitable solution for operational land cover mapping that can cope with the scarcity of in situ data, which is a common challenge in almost every developing country.

Defects are a leading issue for the rejection of parts manufactured through the Directed Energy Deposition (DED) Additive Manufacturing (AM) process. In an attempt to illuminate and advance in situ quality monitoring and control of workpieces, we present an innovative data-driven method that synchronously collects sensing data and AM process parameters with a low sampling rate during the DED process. The proposed data-driven technique determines the important influences that individual printing parameters and sensing features have on prediction at the inter-layer qualification to perform feature selection. Three Machine Learning (ML) algorithms including Random Forest (RF), Support Vector Machine (SVM), and Convolutional Neural Network (CNN) are used. During post-production, a threshold is applied to detect low-density occurrences such as porosity sizes and quantities from CT scans that render individual layers acceptable or unacceptable. This information is fed to the ML models for training. Training/testing are completed offline on samples deemed “high-quality” and “low-quality”, utilizing only features recorded from the build process. CNN results show that the classification of acceptable/unacceptable layers can reach between 90% accuracy while training/testing on a “high-quality” sample and dip to 65% accuracy when trained/tested on “low-quality”/“high-quality” (respectively), indicating over-fitting but showing CNN as a promising inter-layer classifier.

Defects are a leading issue for the rejection of parts manufactured through the Directed Energy Deposition (DED) Additive Manufacturing (AM) process. In an attempt to illuminate and advance in situ quality monitoring and control of workpieces, we present an innovative data-driven method that synchronously collects sensing data and AM process parameters with a low sampling rate during the DED process. The proposed data-driven technique determines the important influences that individual printing parameters and sensing features have on prediction at the inter-layer qualification to perform feature selection. Three Machine Learning (ML) algorithms including Random Forest (RF), Support Vector Machine (SVM), and Convolutional Neural Network (CNN) are used. During post-production, a threshold is applied to detect low-density occurrences such as porosity sizes and quantities from CT scans that render individual layers acceptable or unacceptable. This information is fed to the ML models for training. Training/testing are completed offline on samples deemed &ldquo:high-quality&rdquo: and &ldquo:low-quality&rdquo:, utilizing only features recorded from the build process. CNN results show that the classification of acceptable/unacceptable layers can reach between 90% accuracy while training/testing on a &ldquo:high-quality&rdquo: sample and dip to 65% accuracy when trained/tested on &ldquo:low-quality&rdquo:/&ldquo:high-quality&rdquo: (respectively), indicating over-fitting but showing CNN as a promising inter-layer classifier.

Forests play a central role in the management of the Earth&rsquo:s climate. Airborne laser scanning (ALS) technologies facilitate the monitoring of large and impassable areas and can be used to monitor the 3D structure of forests. While the ALS-based forest measures have been studied in depth, 3D change detection in forests is still a subject of little attention in the literature due to the challenges introduced by comparing point cloud pairs. In this study, we propose an innovative methodology to (i) automatically perform a 3D change detection of forests on an individual tree level: (ii) estimate tree parameters with allometric equations: and (iii) perform an assessment of the aboveground biomass (AGB) variation over time. The area in which the tests were carried out was hit by an ice storm that occurred in the time interval between the two LiDAR acquisitions: furthermore, field measurements were carried out and used to validate the results. The single-tree segmentation of the point clouds was automatically performed with a local maxima algorithm to detect the treetop, and a decision tree method to define the individual crowns around the local maxima. The multitemporal comparison of the point clouds was based on the identification of single trees, which were matched when there was a correlation between the position of the treetops. For each tree, the DBH (diameter at breast height) and the AGB were also estimated using allometric equations. The results are promising and allowed us to identify the uprooted trees and estimate that about 40% of the AGB of the area under examination had been destroyed, with an RMSE over the estimation ranging between 4% and 21% in four scenarios.

Forests play a central role in the management of the Earth’s climate. Airborne laser scanning (ALS) technologies facilitate the monitoring of large and impassable areas and can be used to monitor the 3D structure of forests. While the ALS-based forest measures have been studied in depth, 3D change detection in forests is still a subject of little attention in the literature due to the challenges introduced by comparing point cloud pairs. In this study, we propose an innovative methodology to (i) automatically perform a 3D change detection of forests on an individual tree level; (ii) estimate tree parameters with allometric equations; and (iii) perform an assessment of the aboveground biomass (AGB) variation over time. The area in which the tests were carried out was hit by an ice storm that occurred in the time interval between the two LiDAR acquisitions; furthermore, field measurements were carried out and used to validate the results. The single-tree segmentation of the point clouds was automatically performed with a local maxima algorithm to detect the treetop, and a decision tree method to define the individual crowns around the local maxima. The multitemporal comparison of the point clouds was based on the identification of single trees, which were matched when there was a correlation between the position of the treetops. For each tree, the DBH (diameter at breast height) and the AGB were also estimated using allometric equations. The results are promising and allowed us to identify the uprooted trees and estimate that about 40% of the AGB of the area under examination had been destroyed, with an RMSE over the estimation ranging between 4% and 21% in four scenarios.