Height Model Research Articles

Canopy Segmentation of Overlapping Fruit Trees Based on Unmanned Aerial Vehicle LiDAR

Utilizing LiDAR sensors mounted on unmanned aerial vehicles (UAVs) to acquire three-dimensional data of fruit orchards and extract precise information about individual trees can greatly facilitate unmanned management. To address the issue of low accuracy in traditional watershed segmentation methods based on canopy height models, this paper proposes an enhanced method to extract individual tree crowns in fruit orchards, enabling the improved detection of overlapping crown features. Firstly, a distribution curve of single-row or single-column treetops is fitted based on the detected treetops using variable window size. Subsequently, a cubic spatial region extending infinitely along the Z-axis is generated with equal width around this curve, and all crown points falling within this region are extracted and then projected onto the central plane. The projecting contour of the crowns on the plane is then fitted using Gaussian functions. Treetops are detected by identifying peak points on the curve fitted by Gaussian functions. Finally, the watershed algorithm is applied to segment fruit tree crowns. The results demonstrate that in citrus orchards with pronounced crown overlap, this novel method significantly reduces the number of undetected trees with a precision of 97.04%, and the F1 score representing the detection accuracy for fruit trees reaches 98.01%. Comparisons between the traditional method and the Gaussian fitting–watershed fusion algorithm across orchards exhibiting varying degrees of crown overlap reveal that the fusion algorithm achieves high segmentation accuracy when dealing with overlapping crowns characterized by significant height variations.

Analytic model of outer rib height for thin-walled cylindrical rings in spin-extrusion forming processing

Analytic model of outer rib height for thin-walled cylindrical rings in spin-extrusion forming processing

An Improved Tree Crown Delineation Method Based on a Gradient Feature-Driven Expansion Process Using Airborne LiDAR Data

Accurate individual tree crown delineation (ITCD), which can be used to estimate various forest parameters such as biomass, stem density, and carbon storage, stands as an essential component of precision forestry. Currently, raster data such as the canopy height model derived from airborne light detection and ranging (LiDAR) data have been widely used in large-scale ITCD. However, the accuracy of current existing algorithms is limited due to the influence of understory vegetation and variations in tree crown geometry (e.g., the delineated crown boundaries consistently extend beyond their actual boundaries). In this study, we achieved more accurate crown delineation results based on an expansion process. First, the initial crown boundaries were extracted through watershed segmentation. Then, a “from the inside out” expansion process was guided by a novel gradient feature to obtain accurate crown delineation results across different forest conditions. Results show that our method produced much better performance (~75% matched on average) than other commonly used methods across all test forest plots. The erroneous situation of “match but over-grow” is significantly reduced, regardless of forest conditions. Compared to other methods, our method demonstrates a notable increase in the precisely matched rate across different plot types, with an average increase of 25% in broadleaf plots, 18% in coniferous plots, 23% in mixed plots, 15% in high-density plots, and 32% in medium-density plots, without increasing over- and under- segmentation errors. Our method demonstrates potential applicability across various forest conditions, facilitating future large-scale ITCD tasks and precision forestry applications.

Use of supervised and unsupervised approaches to make zonal application maps for variable-rate application of crop growth regulators in commercial cotton fields

BackgroundZonal application maps are designed to represent field variability using key variables that can be translated into tailored management practices. For cotton, zonal maps for crop growth regulator (CGR) applications under variable-rate (VR) strategies are commonly based exclusively on vegetation indices (VIs) variability. However, VIs often saturate in dense crop vegetation areas, limiting their effectiveness in distinguishing variability in crop growth. This study aimed to compare unsupervised framework (UF) and supervised framework (SUF) approaches for generating zonal application maps for CGR under VR conditions. During 2022–2023 agricultural seasons, an UF was employed to generate zonal maps based on locally collected field data on plant height of cotton, satellite imagery, soil texture, and phenology data. Subsequently, a SUF (based on historical data between 2020–2021 to 2022–2023 agricultural seasons) was developed to predict plant height using remote sensing and phenology data, aiming to replicate same zonal maps but without relying on direct field measurements of plant height. Both approaches were tested in three fields and on two different dates per field.ResultsThe predictive model for plant height of SUF performed well, as indicated by the model metrics. However, when comparing zonal application maps for specific field-date combinations, the predicted plant height exhibited lower variability compared with field measurements. This led to variable compatibility between SUF maps, which utilized the model predictions, and the UF maps, which were based on the real field data. Fields characterized by much pronounced soil texture variability yielded the highest compatibility between the zonal application maps produced by both SUF and UF approaches. This was predominantly due to the greater consistency in estimating plant development patterns within these heterogeneous field environments. While VR application approach can facilitate product savings during the application operation, other key factors must be considered. These include the availability of specialized machinery required for this type of applications, as well as the inherent operational costs associated with applying a single CGR product which differs from the typical uniform rate applications that often integrate multiple inputs.ConclusionPredictive modeling shows promise for assisting in the creation of zonal application maps for VR of CGR applications. However, the degree of agreement with the actual variability in crop growth found in the field should be evaluated on a field-by-field basis. The SUF approach, which is based on plant heigh prediction, demonstrated potential for supporting the development of zonal application maps for VR of CGR applications. However, the degree to which this approach aligns itself with the actual variability in crop growth observed in the field may vary, necessitating field-by-field evaluation.

Optimal integration of forest inventory data and aerial image-based canopy height models for forest stand management

Optimal integration of forest inventory data and aerial image-based canopy height models for forest stand management

Modeling Canopy Height of Forest–Savanna Mosaics in Togo Using ICESat-2 and GEDI Spaceborne LiDAR and Multisource Satellite Data

Quantifying forest carbon storage to better manage climate change and its effects requires accurate estimation of forest structural parameters such as canopy height. Variables from remote sensing data and machine learning models are tools that are being increasingly used for this purpose. This study modeled the canopy height of forest–savanna mosaics in the Sudano–Guinean zone of Togo. Relative heights were extracted from GEDI and ICESat-2 products, which were combined with optical, radar, and topographic variables for canopy height modeling. We tested four methods: Random Forest (RF), Support Vector Machine (SVM), Extreme Gradient Boosting (XGBoost) and Deep Neural Network (DNN). The RF algorithm obtained the best predictions using 98% relative height (RH98). The best-performing result was obtained from variables extracted from GEDI data (r = 0.84; RMSE = 4.15 m; MAE = 2.36 m) and compared to ICESat-2 (r = 0.65; RMSE = 5.10 m; MAE = 3.80 m). Models that were developed during this study can be applied over large areas in forest–savanna mosaics, enhancing forest dynamics monitoring in line with REDD+ objectives. This study provides valuable insights for future spaceborne LiDAR and other remote sensing data applications in similar complex ecosystems and offers local decision-makers a robust tool for forest management.

Naturalness indicators of forests in Southern Sweden derived from the canopy height model

Naturalness indicators of forests in Southern Sweden derived from the canopy height model

Performance of Individual Tree Segmentation Algorithms in Forest Ecosystems Using UAV LiDAR Data

Forests are crucial for biodiversity, climate regulation, and hydrological cycles, requiring sustainable management due to threats like deforestation and climate change. Traditional forest monitoring methods are labor-intensive and limited, whereas UAV LiDAR offers detailed three-dimensional data on forest structure and extensive coverage. This study primarily assesses individual tree segmentation algorithms in two forest ecosystems with different levels of complexity using high-density LiDAR data captured by the Zenmuse L1 sensor on a DJI Matrice 300RTK platform. The processing methodology for LiDAR data includes preliminary preprocessing steps to create Digital Elevation Models, Digital Surface Models, and Canopy Height Models. A comprehensive evaluation of the most effective techniques for classifying ground points in the LiDAR point cloud and deriving accurate models was performed, concluding that the Triangular Irregular Network method is a suitable choice. Subsequently, the segmentation step is applied to enable the analysis of forests at the individual tree level. Segmentation is crucial for monitoring forest health, estimating biomass, and understanding species composition and diversity. However, the selection of the most appropriate segmentation technique remains a hot research topic with a lack of consensus on the optimal approach and metrics to be employed. Therefore, after the review of the state of the art, a comparative assessment of four common segmentation algorithms (Dalponte2016, Silva2016, Watershed, and Li2012) was conducted. Results demonstrated that the Li2012 algorithm, applied to the normalized 3D point cloud, achieved the best performance with an F1-score of 91% and an IoU of 83%.

Shrub height estimation for habitat conservation in NW Iberian Peninsula (Spain) using UAV LiDAR point clouds

ABSTRACT This study aimed to develop and validate a method of estimating 3D parameters from DJI Zenmuse L1 LiDAR on DJI Matrice 300 RTK UAV data to characterise and monitor the structure and conservation status of dense shrub formations. The shrub heights were estimated using Progressive Morphological Filter (PMF) and the Ground Filter module of the FUSION/LDV software. A digital terrain model (DTM) was interpolated with RMSE 0.23 and 0.27 m, respectively, and a normalised canopy height model (nCHM) was calculated by subtracting it from the LiDAR data and the best DTM obtained. The reliability of the estimates was evaluated against georeferenced field data. In addition, the study examined the impact of vegetation characteristics and return reduction in the original point cloud on the accuracy of LiDAR-data derived . Significant differences were found in the correlations between observed and estimated data for the DTM (R2 = 0.9998) and nCHM heights (R2 = 0.51/0.54). The corresponding RMSE values were 0.23 and 0.34 m. Moreover, no significant differences in the reliability were found for different vegetation types, whereas reduction point cloud density (up to 25–50 returns/m2) did not significantly affect accuracy. In conclusion, lightweight UAV LiDAR can effectively detect sub-metric scale vegetation 3D structure, useful for fine-scale habitat conservation.

Developing an AI Tool for Forest Monitoring: Introducing SylvaMind AI

Global forests face increasing threats from deforestation, biodiversity loss, and climate change, necessitating innovative tools for effective monitoring and management. Traditional forest monitoring methods, which rely heavily on manual fieldwork and labor-intensive data processing, are often inadequate for addressing the scale and complexity of these challenges. Advanced tools leveraging artificial intelligence (AI) and remote sensing have emerged as critical solutions, offering timely, accurate, and actionable insights to enable efficient ecosystem monitoring, threat detection, and sustainable management practices. This paper introduces SylvaMind AI, an advanced platform that integrates satellite imagery, deep learning frameworks, and geospatial analysis within a user-friendly interface, which was built using Python for backend systems and deep learning pipelines, alongside tools like Pandas, Rasterio, and TensorFlow for data preprocessing and predictive modelling. The platform processes high-resolution data from Sentinel-2 and Landsat missions for feature extraction and predictive modelling. SylvaMind AI offers two modelling approaches: an automated option for non-technical users and a customizable feature for researchers with specialized needs. Using these approaches, we developed a predictive canopy height model for a study area. The results demonstrated the platform's ability to capture underlying forest patterns and provide detailed insights into canopy height distribution, particularly for medium to high canopies (>25m). This underscores its strength in modeling structural complexity in dense forests. However, the model showed limitations in representing smaller trees, attributed to insufficient training data. SylvaMind AI holds immense potential in transforming forest monitoring by leveraging advanced geospatial data, AI, and intuitive design to address critical challenges in sustainable forest management.

Differences in BMI trajectory for infants in different birth weight classes

Deviations from normal postnatal growth patterns in infancy and early childhood are indicators to predict future negative health outcomes and to direct early interventions, such as changes in feeding regimens. The objective of our study was to examine potential differences in BMI trajectory during the first year of life between infants of different birth weight categories. A pooled, fully anonymized dataset was constructed based on individual subject data of 4,777 participants from 12 clinical studies in which infant growth outcome measurements were collected at several timepoints during their first year of life. A BMI model was constructed using a joint bivariate growth model for weight and height (and their deterministic relationship). A Bayesian approach was used to explore birthweight category-specific differences in BMI development. Our main finding is that the BMI trajectories of the birthweight categories (particularly the small vs large groups) continue to show differences over time, although these differences become less apparent after the first months of age. Both weight and length continue to show consistent differences between categories across the first year of life. A more detailed analysis including the effect of region and gender, showed differential BMI patterns among the three birthweight categories over time across the different subgroups (combinations of region and gender).

Sugarcane Yield Estimation Using UAV-Based RGB Images and Allometric Equations

Accurately estimating pre-harvest sugarcane yield has long been a challenge. There are many methods for predicting sugarcane yield, ranging from simple empirical equations to complex physiological models. This paper reports a study on a method for predicting sugarcane yield using allometric equations combined with UAV-based RGB images (UAVI). UAVI were used for the estimation of the sugarcane height, which is in the form of the sugarcane height model (HM). Sugarcane height is one of the key factors in calculating sugarcane yield in the allometric equation. The results showed that the HM could be used in the allometric equation. There was only a slight discrepancy compared to measurement with a tape measure in the field. In developing the allometric equation, the authors created regression models to estimate aboveground biomass weight in leaf bush (Wl) and millable stalk (Wms) based on sugarcane height (H) and diameter measured from the first segment of sugarcane aboveground (Dfs). The model estimated aboveground biomass weight sufficiently for all stages of cultivation. Based on this model, the authors developed two general equations: Y = 0.0842 H * Dfs0.9827; R2 = 0.93 (used for leaf bush), and Y = 0.1254 H * Dfs1.3926; R2 = 0.93 (used for millable stalk). The decision correlation coefficient (R2) was 80% reliable. The HM models were slightly different from field measurements with a tape measure. Also, there was a little inaccuracy in the root-mean-square error (RMSE) between the sugarcane heights analyzed from UAVI and field measurements using tape measure. The RMSE values, arranged from highest to lowest according to the sugarcane growth stage were: 0.35 m (tillering phase, TP), 0.25 m (grand growth phase, GP), and 0.24 m (ripening phase, RP). Using the HM value in the allometric equation effectively estimated the sugarcane yield at different growth stages. Sugarcane growth at TP and GP phases gave the highest sugarcane yield at 200 cm of the HM value, with total yields of 11.49 and 16.75 ton/hectare, respectively. Whereas, RP gave the highest sugarcane yield at 350 cm of the HM value (total yield at 42.97 ton/hectare). Overall, an accurate estimation of the aboveground biomass of leaf bush and millable stalk can be obtained using these equations. The authors believe that the research methods presented here can help sugarcane farmers to better estimate yield prior to harvest.

A novel intensity-based phase unwrapping method for highly discontinuous topology

Abstract The research on the spatial phase unwrapping method is primarily concerned with unwrapping the phase of highly discontinuous objects. This study presents a phase unwrapping method that utilizes an illumination attenuation model to effectively unwrap the phase of discontinuous objects. Initially, the attenuation features of light intensity with distance are examined, and using the phase height model in the fringe projection profilometry system, the unwrapped light intensity-phase difference distribution map is computed. Next, the phase mask is utilized to divide the non-continuous area, and the process of phase unwrapping is carried out on each individual section. Ultimately, the phase jump order produced by the light intensity-phase difference is employed to direct the sub-regions in accomplishing the phase fusion and creating the global unwrapped phase map.The method shown in this study accomplishes phase unwrapping in areas with discontinuities without adding complexity to the system or requiring additional projection images. Furthermore, it is also successful for isolated regions with phase jumps. This research verifies the efficacy of the phase unwrapping algorithm in handling phase discontinuity through simulation and experimentation.

Enhancing wave energy farm efficiency: Eigen-stacking ensemble framework

The significant wave height (SWH) plays a pivotal role across diverse domains, ranging from the assessment of wave energy potential to the optimization of marine operations and the advancement of ocean engineering techniques. Understanding the spatial and temporal distribution patterns of SWH data among buoy network stations holds utmost importance for various applications. This research introduces a novel methodology for accurate spatiotemporal SWH prediction across multiple stations within buoy monitoring networks. Leveraging eigen SWH time series enables the identification of dataset patterns, feature extraction, and dimensionality reduction. Examining eigen time series provides insights into SWH dynamics, allowing exclusive use for prediction across all stations. By incorporating this eigen SWH time series as the sole input into a stacking ensemble model, we introduce a highly efficient and accurate framework for SWH prediction. Notably, the methodology achieves success in forecasting hourly SWH values along the western United States coastline, utilizing a stacking ensemble technique for enhanced accuracy. Evaluation metrics confirm outstanding performance, with CE and R2 values surpassing 0.95 and 0.93, respectively. Further, a single Long Short-Term Memory (LSTM) model is incorporated as a benchmark to assess the effectiveness of the stacking ensemble model for significant wave height (SWH) prediction, with results demonstrating that the ensemble model outperforms the single LSTM in terms of accuracy and robustness. Consequently, the introduced stacking ensemble model promises autonomy in forecasting spatial-temporal SWH patterns, extending its applicability within ocean engineering and scientific research.

Study on the predictive model for continuous build height of 3D printed concrete (3DPC) based on printability and early mechanical properties

Study on the predictive model for continuous build height of 3D printed concrete (3DPC) based on printability and early mechanical properties

Forest stand height predicted from ICESat-2 ATLAS data for forest inventory and comparison to airborne laser scanning metrics

Abstract The study analysed 2019–2022 summertime canopy height predictions (H ICESat) given in ICESat-2 ATLAS dataset ATL08 for hemiboreal forests growing on an area of 40,000 km2 in Estonia around 25.6° E, 58.8° N. In total 12,711 ATL08 20×20 m pixel observations were used from 3,065 forest stands with homogenous canopy structure. Regression modelling was used to explain variability in ground surface elevation estimates, and relationships of H ICESat to basal area weighted mean tree height given in the forest inventory database (H FI) and to the 95th percentile of the vertical distribution of airborne laser scanning pulse return (H ALS). The other explanatory variables were the ICESat-2 ATLAS observation geographic location, ICESat-2 ATLAS track and beam energy indicators, forest canopy cover, evergreen coniferous tree dominance indicator, and deep peat soil indicator. The linear model between the Estonian digital terrain model elevation and ATL08 ground elevation had a determination coefficient of R2=99.97% and residual standard error of δ=0.51 m when a geographic location was included. The H FI can be predicted from H ICESat with R2=85% and δ=2.7 m. A comparison of means indicated that, on average, H ICESat was about 0.3 m greater than H FI. All the predictive variables (except the geographic location) were significant in canopy height models, and the best models fitted H ICESat with R2=95% and δ=1.6 m, however, there was no notable increase in R2 if more predictors than H ALS were added in the models. In practical applications using ATL08 data for forest inventories, the inclusion of weak energy beam observations increases the number of observations, but the beam energy indicator has to be included in the models.

Research on Strip Coal Pillar Technology under the Construction Group of Gangue Filling and Recycling

In order to reduce the various impacts of surface movement caused by coal mining on the human living environment, avoid the accumulation of coal gangue on the surface, and recover the strip coal pillar resources left under urban areas, based on the stability of the binary bearing structure composed of backfilled gangue and narrow coal pillars after strip mining and the characteristics of the equivalent mining height model, theoretical analysis and numerical simulation methods were comprehensively used to analyze the factors affecting the stability of coal pillars, the ultimate strength of coal pillars, and the safety factor of coal pillar stress under the condition of binary bearing structure. Through comparative analysis of deformation of surrounding rock after strip mining and coal pillar recovery, it was found that the mining pressure manifestation of strip coal pillar gangue filling is weak, without obvious initial and periodic pressure phenomena, and the top control is relatively easy without the risk of dynamic pressure. The research results provide useful reference and guidance for the first application of fully mechanized mining technology in urban strip coal pillar recovery in coal mines.

Study on the flame behavior of ethanol spill fire on base plate with different width in longitudinal ventilation environment

The intricate flame behavior of spill fire in a ventilated tunnel significantly aggravates the severity of accident consequences. Flame height, flame tilt angle, and combustion area were measured by conducting ethanol spill fire experiments in model tunnel with different wind speeds in base plates of different width. Furthermore, critical transition wind speed, dimensionless analysis of burning rate and heat transfer analysis were performed. The findings reveal that, in windy conditions, wind speed linearly lowers flame height, while baseplate width’s impact is irregular. Longitudinal ventilation notably causes flame inclination, base plate width and wind speed jointly affect inclination at low wind speed, however, in high wind speeds, the inclination is mainly determined by the wind speed. With the increase of wind speed, the burning rate firstly increases and then decreases, resulting in an initial decline followed by an increase in the combustion area. The increase in base plate width not only leads to a continuous increase in the combustion area, but also results in a decrease of the critical transition wind speed at which the burning rate changes. An equation for the dimensionless burning rate in upwind environments has been developed, revealing that burning rate is significantly higher in longitudinal wind conditions compared to still air. Furthermore, the difference becomes more pronounced as the baseplate width decreases. By taking into account the asymmetric air entrainment of the spill fire flame, a revised model for flame height and inclination has been introduced, utilizing a dimensionless array (2n+1/n). This model effectively predicts the morphological changes of spill fire flames under windy conditions, and the dimensionless flame height of spill fires is approximately 0.18 to 0.36 times that of pool fires.

Detection of snow disturbances in boreal forests using unitemporal airborne lidar data and aerial images

Abstract Snow is among the most significant natural disturbance agents in Finland. In silviculture, maps of snow disturbance are needed to recognize severely disturbed forests where the risk of subsequential disturbances, such as insect outbreaks, is high. We investigated the potential of unitemporal airborne lidar (light detection and ranging) data and aerial images to detect snow disturbance at the tree level. We used 81 healthy and 128 snow-disturbed field plots established in a 63 800 ha study area in Eastern Finland. A subset of trees (n = 675) was accurately positioned in the field plots. We carried out individual tree detection (ITD) using airborne lidar data (5 p/m2), and a random forest classifier was used to classify healthy and broken trees. Tree features were extracted from a terrain elevation model, lidar data, and aerial imagery. We compared canopy height model–based (ITDCHM) and point cloud–based (ITDPC) ITD approaches. We explored random forest variable importance scores and evaluated the classification performance by an F1-score and its components (precision and recall). Performance was also evaluated at the plot level to investigate errors associated with the predicted number of broken trees. We achieved F1-scores of 0.66 and 0.85 for the tree- and plot-level classifications, respectively. The variable importance scores showed that elevation above sea level was the most important predictor variable followed by ITD-based features characterizing the neighborhood of trees. The ITDCHM slightly outperformed the ITDPC at the tree level, while they both underestimated the number of broken trees at the plot level. The proposed approach can be carried out alongside lidar-assisted operational forest management inventories provided that a set of positioned broken and healthy trees are available for model training. Since airborne lidar data often have a temporal resolution of several years for the same areas, future research should consider the utilization of other remotely sensed data sources to improve the temporal resolution.

Derivation of surface models using satellite imagery deep learning architectures with explainable AI

Global mapping of urban morphology and human settlement is an area of great interest, with significant scientific and humanitarian value. Current approaches for high-resolution surface modeling using aerial LiDAR are impressive and useful, but it is impractical for application on a global scale. This is due to technical challenges (efficiency of coverage, etc), as well as cost. These are significant obstacles, particularly in the less wealthy regions of the world, many of which are experiencing the greatest increase in population and physical infrastructure. Advances in satellite imagery and deep learning are now enabling the lower resolution alternative, but one which is more scalable with global application. This paper examines the efficacy of four state-of-the-art deep learning architectures (DeepLabv3+, SegNet, U-Net, and U-Net++) for application to Digital Surface Modeling (DSM), particularly for urban districts. The study considered New York City as the test area, using satellite based Radar backscatter images and visible and near infrared images for surface model development, along with high resolution LiDAR based surface height model of the city. We performed a quantitative evaluation using metrics such as Root Mean Squared Error (RMSE), Mean Absolute Error (MAE) and R-Squared, and a qualitative analysis to identify the best performing model. U-Net++ proved to be the most effective, with superior accuracy in capturing key urban features. In addition, we employed Explainable Artificial Intelligence (XAI) techniques, such as Gradient-Weighted Class Activation Mapping (GRAD-CAM) and Saliency Maps, to improve model interpretability, revealing important information about decision-making processes and highlighting critical input features. The results of this study contribute to the advancement of remote sensing techniques for urban analysis, with potential applications. Despite the overall success, the study also identified challenges, such as the loss of detailed building shapes in the predicted DSMs and the influence of optical shadowing and SAR-induced quiescence on model performance.