Forest conservation and management are increasingly challenged by evolving societal expectations, biodiversity decline, and the impacts of climate change, requiring accurate, spatially detailed data for effective decision-making. Remote sensing and photogrammetry have become critical tools allowing detailed mapping and measurement of forests worldwide. While traditional satellite and airborne remote sensing remains important, ground-based data is becoming increasingly important for monitoring biodiversity and optimising management. Despite advances in deep learning for tree segmentation and species classification, the lack of extensive, high-quality labelled datasets is hampering development. To address this issue, TreeScanPL10K is being introduced - a dataset that surpasses previous resources in scope, comprising 10,417 individual trees, with species identified for approximately 72%. The dataset, which represents most of the forest-forming species in Central Europe, was collected in various Polish forest stands of different ages, composition and development stages. TreeScanPL10K aims to support researchers and forestry professionals by enabling the training and testing of advanced analytical tools, promoting transparency and accelerating progress in precision forestry and ecological studies.

ABSTRACT Moisture is a critical factor threatening the durability of architectural heritage, particularly masonry and stone structures. This paper presents an innovative moisture-protection solution implemented at the historic Cistercian monastery in Gdańsk Oliwa, involving the construction of a reinforced-concrete drying wall combined with a base slab and a covered excavation. The construction methods and applied materials are described in detail, and the effectiveness of the solution is evaluated using quantitative moisture measurements conducted before and after the intervention. Terrestrial laser scanning (TLS) was applied throughout all construction stages, providing high-resolution spatial documentation and monitoring of geometric conditions. Complementary ground-penetrating radar (GPR) surveys supported diagnostics by identifying hidden defects, subsurface infrastructure, and assessing structural integrity. The integration of TLS and GPR technologies enabled an accurate assessment of the technical conditions and quality of workmanship, which significantly improved the protection of the monastery’s foundations against moisture and contributed to the long-term safety and preservation of the historic structure.

Cross-domain semantic classification of 3D point clouds remains challenging due to strong domain shifts between heterogeneous sensing modalities. Most existing classification frameworks are domain-specific, limiting their use in integrated land–water mapping applications. This study evaluates the transferability of classical geometric machine learning classifiers between terrestrial and underwater point cloud domains without target-domain retraining. Experiments were conducted using terrestrial data acquired with a Leica BLK360 terrestrial laser scanner (TLS) and underwater point clouds collected with a Blueview BV5000 mechanical scanning sonar (MSS). Two dimensionality-based frameworks, CANUPO–Support Vector Machine (SVM) and 3DMASC–Random Forest (RF), were implemented in CloudCompare and assessed under intra-domain and cross-domain configurations. Strong intra-domain performance was achieved, with terrestrial–terrestrial accuracies of 0.99 for CANUPO–SVM and 0.97 for 3DMASC. In underwater evaluation, CANUPO maintained high accuracy (0.97), whereas 3DMASC decreased to 0.86 due to increased variability in the submerged data. Under cross-domain transfer, CANUPO achieved 0.93 accuracy for terrestrial-to-underwater and 0.89 for underwater-to-terrestrial classification, while 3DMASC demonstrated stable generalisation with 0.95 accuracy in both directions. Overall, dimensionality-based geometric descriptors capture stable structural cues across sensing environments, providing an interpretable and efficient pathway for applications such as hydrographic surveying, coastal monitoring, and underwater search-and-rescue detection. Future work will extend validation to larger datasets and explore domain adaptation strategies to further reduce cross-modality domain shift.

This study validates terrestrial laser scanning (TLS) data by comparing it with traditional geodetic levelling in a bridge pavement survey. Bridge roadway slopes were determined independently using both geodetic levelling and TLS. The comparison demonstrated that differences between slope values obtained by the two methods do not exceed 0.1 ‰, with a standard deviation of 0.05, indicating a high level of agreement and reliability of TLS data. The findings confirm the practical benefits of integrating terrestrial laser scanning into infrastructure inspection technology to enhance the reliability and effectiveness of structural monitoring.

Erosion rates in rock salt exposures with diverse karren monitored by erosion pins, close-range photogrammetry and terrestrial laser scanner

Calibrating inter-target distances in a network with high accuracy using a terrestrial laser scanner

High terrestrial laser scanning: An accessible method for tree crown characterization

Many important processes in forest ecosystems are influenced by the spatial structure of the canopy. The spatial arrangement of trees and their branches within the canopy is crucial for light interception, often resulting in a positive relationship between canopy space filling and stand productivity. To date, there is no universal definition of canopy space. However, different parts of the canopy have distinct characteristics, making it crucial to assess the significance of varying canopy definitions. In this study, we investigated how canopy space filling changes with four different definitions of canopy space by using a canopy space filling index (CSFI) derived from terrestrial laser scanning. Moreover, we assessed the relative importance of these definitions in explaining stand productivity. Using data from a tropical tree diversity experiment, we found that CSFI strongly varied with the specific delineation of the canopy space. Across canopy definitions, stand productivity was significantly and positively affected by CSFI, indicating that this relationship was generally robust to varying definitions of canopy space. However, excluding the uppermost part and including the lowest part of the canopy space resulted in a distinct decline of explained variance in productivity. In addition, CSFI played a more important role in regulating productivity in mixtures compared to monocultures. Our results highlight the importance of different layers of the canopy space and tree species richness for a better understanding and assessment of forest dynamics and ecosystem functions. Further research is needed on the relationship between canopy space filling and stand productivity based on different canopy space definitions for mature forests and from different biomes.

Uncovering growth dynamics in Scots pine through the detection of the onset of height growth using near-daily terrestrial laser scanning time series

The use of a Terrestrial Laser Scanner (TLS) to create terrain profiles and scan for opencast mines still relies on Ground Control Points (GCPs) for geo-referencing. The use of GCPs has several limitations related to the time and cost required for field measurements. Besides this, not all locations are suitable for GCP measurements due to poorly accessible terrain or security reasons. Direct geo-referencing technique for determining precise reflector target and TLS position and orientation in the TLS utilizing Global Navigation Satellite System (GNSS) with RTK methodology. This study aims to assess the backsight distance of the reflector target from the TLS in direct geo-referencing using precise coordinates of GNSS. In this study, the backsight distance result did not yield a positive correlation with its accuracy. A t-test has been done between TLS and Reflector distance and Delta. The t-test for statistical significance of the data found that the level of significance (P) is less than 0.0001. It shows statistically significant data at the confidence level of 95 %. The combination of GNSS and TLS is used in terms of accuracy and time, as GNSS helps in maintaining the precise coordinates, and the TLS scan covers larger-scale mapping activities like topographically survey.

Abstract Background Weather and fuels are among the critical, interacting factors that drive wildland fire behavior, and thus are primary factors in fire operations planning and decision support tools. In mesic forests, variation in stand structure may lead to heterogeneous microclimate and fuel moisture conditions in the understory where fires often ignite and spread. However, such variation in fuel availability is often overlooked by fire behavior models that assume spatially uniform weather and fuel conditions. In this study, we analyze a combination of field-based understory meteorology and fuel moisture data with terrestrial laser scanning (TLS) and traditional forest inventory data to develop new knowledge about relationships between forest structure, microclimate, and dead fuel moisture in USA northern conifer forests that can inform fire operations planning and decision support. Results We found that open canopy plots were significantly warmer (+ 7.82 °C average daily maximum air temperature) and drier (-24.1% average daily relative humidity) and with drier fuels (+ 8.03% fuel moisture) at midday in summer compared to closed canopy plots. Directly using microclimate variables (i.e., air temperature and relative humidity) resulted in better predictions of dead fuel moisture content (mean R 2 = 0.88) than using forest structure variables such as canopy openness (R 2 = 0.60). Furthermore, forest structure variables derived from TLS were better predictors of dead fuel moisture content (R 2 = 0.74) than traditional forest inventory metrics. Conclusions Our study used multi-modal measurements to demonstrate that dense forest cover linearly reduces fuel availability by buffering microclimate and maintaining fuel moisture. This research can be used to develop thinning prescriptions to achieve certain thresholds of understory temperature, relative humidity, and dead fuel moisture. Moreover, our results highlight the microclimate buffering effect of shaded fuel breaks used in fire suppression and containment tactics. Finally, our work suggests that tools like TLS can be used to fine tune fuel-weather relationships in fire behavior models that use spatially explicit fuels data to inform planning and predict fuels treatment effectiveness. This research enhances fire managers’ ability to plan and implement fuel treatments by highlighting how changes in forest stand structure drive fine scale heterogeneity in fuel availability.

Key message Aboveground biomass (AGB) increased from forest edge to forest interior in small forest patches of Western-Africa. In plots of 0.25 ha, AGB did not correlate with tree species richness or wood density. AGB in these unprotected forest patches was lower than in the nearby protected forests. AGB obtained from manual inventory and terrestrial laser scanning correlated moderately. Context In Western-Africa, small, unprotected forest patches amidst agricultural lands provide vital ecosystem services like carbon storage. However, accurately measuring aboveground biomass remains challenging, and terrestrial laser scanning (TLS) might become an accurate, non-destructive method. Aims This study explored AGB, its spatial distribution, and relationships with ecological determinants, and compared AGB estimated from manual inventory with that from TLS. Methods We established 109 plots and inventoried 9591 trees across seven forests in Togo, Benin, Nigeria, and Cameroon. AGB was obtained from allometric equations using diameter and tree heights, as well as from segmented point-clouds. Plot-level AGB was extrapolated to the entire forest. Results AGB in forest patches ranged from 85 to 259 Mg/ha, which is lower than in nearby protected forests. Forests close to the equator have generally higher AGB, and most forests showed reduced AGB and wood density close to forest edges. AGB showed no correlation with wood density, structural complexity, and tree species richness. AGB estimations by manual inventory and TLS correlated moderately. Conclusion Our findings highlight the value of ground-based methods and the need to connect and protect forests as carbon reservoirs.

Abstract Quantifying the distribution of Spanish moss ( Tillandsia usneoides L.) is challenging because it grows suspended from high tree branches, limiting manual sampling. Terrestrial laser scanning (TLS) provides a non-destructive means of capturing vegetation structure in three dimensions. However, no established methods exist for identifying Spanish moss from TLS data. We evaluated five classification methods for distinguishing Spanish moss in TLS-derived point cloud data: Graph, DBSCAN, Random Forest (RF), Kernel Point Convolution (KPConv), and PointNet++. PointNet++ achieved the highest accuracy (81%), followed by DBSCAN (70%), KPConv (61%), RF (54%), and Graph (52%). Unsupervised methods required minimal computational resources (2–3 min, 8–16 GB memory) without training. RF required 3 h for training, 8 for prediction with 1024 GB memory. Deep learning methods required substantially more: KPConv needed 60 h for training, 4 for prediction (256 GB), while PointNet++ required 48 h for training, 1 for prediction (128 GB). Agreement was lowest in the central and upper canopy due to occlusion. Surface variation, PCA1, and verticality contributed most to accurate predictions. These results demonstrate the feasibility of using TLS and advanced classification methods for non-destructive Spanish moss mapping and highlight the accurate classification ability of PointNet++ for future biomass estimation at landscape scales.

This study assesses transformer-based 3D semantic segmentation models for detecting structural components in terrestrial laser scans, given that no training data currently exists for shell construction sites. Manual annotation of 3D point clouds is expensive, yet high-quality labels remain essential for supervised computer vision and validation. Automated pre-labeling can cut down annotation effort by shifting human tasks from exhaustive labeling to targeted verification and correction, assuming models can robustly identify the most common structural elements. We designed a three-stage evaluation protocol covering (i) supervised learning, (ii) cross-domain generalization, and (iii) transfer learning with limited labeled data in the target domain to test model generalization in this context. Three transformer architectures (Point Transformer V2, Point Transformer V3, and Swin3D) are evaluated using four established indoor datasets (S3DIS, ScanNetV2, Structured3D, and VASAD) and a custom domain-specific dataset of annotated construction scenes. Training only on the limited construction dataset results in weak generalization. In contrast, pretraining on loosely related synthetic data and fine-tuning on a minimal number of labeled construction scenes enable reliable segmentation of core building components. A sensitivity analysis also showed that just 12 samples are sufficient to calibrate a pretrained model to a specific building type. The models perform well despite differences between synthetic training data and noisy real-world scans. Among the evaluated architectures, Swin3D delivers the best performance, with +18% mIoU improvement through general pretraining, while PTv3 converges faster with fewer target-domain samples. These findings suggest that transfer learning with limited labeled construction data offers a practical foundation for scalable pre-labeling workflows and human-in-the-loop applications in architecture, engineering, and construction.

ABSTRACT The growth process shapes the structure of trees in forest, and the rate the stems thicken over time is linked with wood properties such as ring width and wood density. This study investigated the potential of bi-temporal terrestrial laser scanning (TLS) for measuring the secondary growth of coniferous trees and assessing wood properties derived from core samples using X-ray microdensitometry over a nine-year period in boreal forests. TLS-derived DBH measurements demonstrated high agreement with reference data (r > 0.96). Volumetric secondary growth ( Δ v ) slightly outperformed basal area increment ( Δ g ) in capturing secondary growth dynamics as moderate correlations were found between TLS-derived estimates and mean ring width (RWm; r = 0.60–0.67). In contrast, correlations with ring basal-area-weighted wood density (WDg) were weak but statistically significant (r = –0.16 to –0.18) for both TLS and core-sampled measurements. The findings suggest that TLS can be used to measure secondary growth, but the ability to predict wood density of coniferous species via secondary growth measurements – neither with TLS nor core-sampling – is limited due to internal anatomical factors not captured by external measurements. Overall, these findings support the integration of TLS into forest monitoring frameworks, as it provides equally reliable yet more versatile data than callipers for measuring secondary growth.

Terrestrial Laser Scanning (TLS) can provide detailed three-dimensional structural information for individual trees and has become an important data source for tree species classification. However, most existing models are trained using leaf-on point clouds and therefore tend to rely heavily on leaf distribution and crown appearance. When the input changes from leaf-on point clouds to woody-dominated representations, classification performance often declines. To address this issue, this study proposes a mixed-input tree species classification framework for six typical temperate broadleaf tree species. First, a KPConv-based wood–leaf separation model was used to extract woody point sets from leaf-on TLS point clouds, thereby generating woody-only representations for subsequent classification. Second, a multi-task learning network based on DGCNN was constructed. In addition to the main task of tree species classification, an auxiliary task for input-representation discrimination was introduced to enhance the model’s adaptability to different input forms. Experiments were conducted using a dataset composed of local TLS samples from China and publicly available single-tree point clouds from the BioDiv dataset. The results show that the proposed method achieved an overall accuracy of 94.3% on the mixed test set of six typical broadleaf tree species, with average Precision, Recall, and F1 values of 94.3%, 93.6%, and 93.9%, respectively. These results indicate that integrating woody structural representations with multi-task learning can effectively alleviate overreliance on leaf-on appearance features and improve classification robustness under different input representations.

The Indus Valley Civilization, one of the oldest and most diverse heritages, must be meticulously preserved. We aim to utilize modern techniques and tools to conserve these extensive built heritage sites for the benefit of society and future generations. Accurate digitization through non-invasive and long-range imaging tools is essential for preserving and showcasing India's heritage on the global platform. These advanced methods capture intricate details, ensuring precise documentation of geometric, structural, and architectural elements, thereby maintaining their historical value. In this study, we propose a novel approach to document a 125-year-old Indo-Saracenic building complex in the Madras High Court campus using advanced reality capture techniques, such as close-range photogrammetry and terrestrial laser scanning surveys. The three-dimensional model developed from these hybrid techniques enables as-built documentation and precise measurement. They also support the extraction of 2D drawings, facilitating the planning and execution of restoration works. Furthermore, the developed model is integrated with virtual reality techniques to create an immersive and walk-through model. The outcome of our work is an application that offers a virtual and interactive environment with digital datasets supporting the conservation, restoration, and management of built heritage structures and also promotes digital heritage tourism in the near future.

Accurate 3D documentation of large and complex structures is essential for long-term stability assessment, structural monitoring, and conservation planning, particularly for heritage sites exposed to environmental and anthropogenic threats. This study develops an integrated workflow combining Terrestrial Laser Scanning (TLS), Global Navigation Satellite System (GNSS), and Total Station geodetic control for large-scale, high-precision documentation. The approach was implemented at the Saqqara archaeological zone, a UNESCO World Heritage Site facing significant deterioration risks, to document four major pyramids: Djoser, Unas, Teti, and Userkaf. More than 2.1 billion georeferenced points were acquired from 16 scan positions with sub-centimeter registration errors and overall geometric accuracy better than ±1 cm. From these datasets, detailed mesh models, orthoimages, Digital Elevation Models (DEMs), contour maps, and 2D plans were derived. These enabled quantitative analyses of height loss and volumetric change, indicating severe structural degradation in Unas (~53%), Teti (~66%), and Userkaf (~63%), as well as localized deformations such as 4.2 cm displacement at Teti’s south flank. The degradation results from environmental factors and anthropogenic influences. Beyond this case study, the workflow proves that integrated TLS documentation can be applied to large and complex structures, supporting deformation monitoring, stability assessment, and digital twin development.

Unlike image-only systems that falter in shadows, glare, and low contrast, LiDAR directly records surface geometry and supports depth-aware quantification. This survey examines LiDAR-based road surface damage classification across the entire pipeline, encompassing acquisition with mobile and terrestrial laser scanning, preprocessing and representation choices, supervised, semi-supervised, and unsupervised learning techniques, as well as multisensor fusion at early, mid, and late stages. A consistent thread is measurement, not just detection: we describe how LiDAR damage classification maps to agency practices such as the Distress Identification Manual and the Pavement Condition Index. We summarize datasets and evaluation protocols for detection, segmentation, 3D reconstruction, and ride quality. We outline practical concerns for corridor-scale deployment: calibration and timing, intensity normalization, tiling/streaming, and runtime budgeting. The review concludes with open problems and outlines directions for robust, severity-aware, and scalable field systems.

ABSTRACT Terrestrial laser scanning observation network (TLS-ON) planning for large-scale building façades faces a challenge in balancing the data coverage and deployment cost. Local optimization methods are computationally efficient, but they often generate redundant scan positions. However, global optimization methods suffer from a high computational cost and a reliance on a high-quality initial population. In this paper, we propose a novel method integrating local and global optimization strategies for TLS-ON planning with high efficiency and quality. First, a greedy algorithm generates a locally optimal solution. A Gaussian kernel function is then developed to construct a high-quality and diverse initial population, based on this local solution. Finally, the constrained non-dominated sorting genetic algorithm II (NSGA-II) combined with the Pareto-Edgeworth-Grierson (PEG) method is applied to obtain a globally optimal network. Experimental validation at a large-scale campus site in China demonstrates that the proposed method reduces scan positions by 13.8% and improves coverage by 2.41%, compared to local optimization, while accelerating convergence by a maximum of 5.5 times over global optimization with random initialization. Real scanning cases confirm its practicality, achieving high coverage in field validation. This work offers an efficient solution for 3D data acquisition in applications such as building information modeling and urban digital twins.