Comparative Analysis of Algorithms to Cleanse Soil Micro-Relief Point Clouds DOI Creative Commons
Simone Ott, Benjamin Burkhard,

Corinna Harmening

et al.

Geomatics, Journal Year: 2023, Volume and Issue: 3(4), P. 501 - 521

Published: Nov. 26, 2023

Detecting changes in soil micro-relief farmland helps to understand degradation processes like sheet erosion. Using the high-resolution technique of terrestrial laser scanning (TLS), we generated point clouds three 2 × 3 m plots on a weekly basis from May mid-June 2022 cultivated Germany. Three well-known applications for eliminating vegetation points cloud were tested: Cloth Simulation Filter (CSF) as filtering method, variants CANUPO machine learning and ArcGIS PointCNN deep sub-category using neural networks. We assessed methods with hard criteria such F1 score, balanced accuracy, height differences, their standard deviations reference surface, resulting data gaps robustness, soft time-saving capacity, accessibility, user knowledge. All algorithms showed low performance at initial measurement epoch, increasing later epochs. While most results demonstrate better PointCNN, this algorithm revealed an exceptionally plot 1, which is describable by generalization gap. Although created highest amount gaps, recommend that include colour values combination CSF.

Language: Английский

Aboveground biomass modeling using simulated Global Ecosystem Dynamics Investigation (GEDI) waveform LiDAR and forest inventories in Amazonian rainforests DOI
Nadeem Fareed, Izaya Numata, Mark A. Cochrane

et al.

Forest Ecology and Management, Journal Year: 2025, Volume and Issue: 578, P. 122491 - 122491

Published: Jan. 5, 2025

Language: Английский

Citations

4
A Comprehensive Review of LiDAR Applications in Crop Management for Precision Agriculture DOI Creative Commons
Sheikh Muhammad Farhan, Jianjun Yin, Zhijian Chen

et al.

Sensors, Journal Year: 2024, Volume and Issue: 24(16), P. 5409 - 5409

Published: Aug. 21, 2024

Precision agriculture has revolutionized crop management and agricultural production, with LiDAR technology attracting significant interest among various technological advancements. This extensive review examines the applications of in precision agriculture, a particular emphasis on its function cultivation harvests. The introduction provides an overview highlighting need for effective growing significance technology. prospective advantages increasing productivity, optimizing resource utilization, managing diseases pesticides, reducing environmental impact are discussed. comprehensively covers detailing airborne, terrestrial, mobile systems along their specialized field. After that, paper reviews several uses cultivation, including growth yield estimate, disease detection, weed control, plant health evaluation. use soil analysis management, mapping categorization measurement moisture content nutrient levels, is reviewed. Additionally, article how used harvesting crops, autonomous systems, post-harvest quality evaluation, prediction maturity yield. Future perspectives, emergent trends, innovative developments discussed, critical challenges research gaps that must be filled. concludes by emphasizing potential solutions future directions maximizing LiDAR’s agriculture. in-depth gives helpful insights academics, practitioners, stakeholders interested using this environmentally friendly which will eventually contribute to development methods.

Language: Английский

Citations

13
Towards intelligent ground filtering of large-scale topographic point clouds: A comprehensive survey DOI Creative Commons
Nannan Qin, Weikai Tan, Haiyan Guan

et al.

International Journal of Applied Earth Observation and Geoinformation, Journal Year: 2023, Volume and Issue: 125, P. 103566 - 103566

Published: Nov. 21, 2023

With the fast development of 3D data acquisition techniques, topographic point clouds have become easier to acquire and promoted many geospatial applications. Ground filtering (GF), as one most fundamental challenging tasks for post-processing large-scale clouds, has been extensively studied but yet be well solved. To reveal future superior solutions, a comprehensive investigation up-to-date GF studies is essential. However, existing surveys are scarce fail capture latest progress advancements. this end, paper not only presents review advanced methods, also conducts systematic comparative analyses experimental results on public benchmark datasets. Moreover, survey compiles recent publicly available resources that can leveraged research, including pertinent datasets, metrics, range open-source tools. Finally, remaining challenges promising research directions GF, implications understanding, discussed in-depth. It expected simultaneously serve position tutorial those interested in GF.

Language: Английский

Citations

17
UAS Quality Control and Crop Three-Dimensional Characterization Framework Using Multi-Temporal LiDAR Data DOI Creative Commons
Nadeem Fareed, Anup Das, Paulo Flores

et al.

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(4), P. 699 - 699

Published: Feb. 16, 2024

Information on a crop’s three-dimensional (3D) structure is important for plant phenotyping and precision agriculture (PA). Currently, light detection ranging (LiDAR) has been proven to be the most effective tool crop 3D characterization in constrained, e.g., indoor environments, using terrestrial laser scanners (TLSs). In recent years, affordable onboard unmanned aerial systems (UASs) have available commercial applications. UAS (ULSs) recently introduced, their operational procedures are not well investigated particularly an agricultural context multi-temporal point clouds. To acquire seamless quality clouds, ULS parameter assessment, flight altitude, pulse repetition rate (PRR), number of return echoes, becomes non-trivial concern. This article therefore aims investigate DJI Zenmuse L1 practices traditional density, canopy height modeling (CHM) techniques, comparison with more advanced simulated full waveform (WF) analysis. Several pre-designed flights were conducted over experimental research site Fargo, North Dakota, USA, three dates. The altitudes varied from 50 m 60 above ground level (AGL) along scanning modes, repetitive/non-repetitive, frequency modes 160/250 kHz, echo (1n), (2n), (3n), assessed diverse dry corn, green sunflower, soybean, sugar beet, near harvest yet changing phenological stages. Our results showed that mode (2n) captures better than (1n) (3n) whereas provides highest penetration at 250 kHz compared 160 kHz. Overall, CHM heights correlated situ measurements R2 (0.99–1.00) root mean square error (RMSE) (0.04–0.09) m. Among all crops, soybeans lowest correlation (0.59–0.75) RMSE (0.05–0.07) We weaker occurred due selective underestimation short crops influenced by phonologies. explained mode, PRR, analysis unable completely decipher impact acquired For first time context, we phenology meaningful clouds revealed WF analyses. Nonetheless, present study established state-of-the-art benchmark framework optimization datasets.

Language: Английский

Citations

4
Package Positioning Based on Point Registration Network DCDNet-Att DOI Open Access
Juan Zhu, Chunrui Yang,

Guolyu Zhu

et al.

Electronics, Journal Year: 2025, Volume and Issue: 14(2), P. 352 - 352

Published: Jan. 17, 2025

The application of robot technology in the automatic transportation process packaging bags is becoming increasingly common. Point cloud registration key to applying industrial robots systems. However, current point models cannot effectively solve deformed targets like bags. In this study, a new network, DCDNet-Att, proposed, which uses variable weight dynamic graph convolution module extract features. A feature interaction used common features between source and template cloud. same geometric two pairs clouds are strengthened through bottleneck module. channel attention model obtain weights. each spatial position calculated, rotation translation structure sequentially quaternions vectors. fitting loss function constrain parameters neural network have larger receptive field. Compared with seven methods, including ICP algorithm, GO-ICP FGR proposed method had errors (MAE, RMSE, Error 1.458, 2.541, 1.024 ModelNet40 dataset, respectively) 0.0048, 0.0114, 0.0174, respectively). When registering dataset Gaussian noise, Error) were 2.028, 3.437, 2.478, respectively, 0.0107, 0.0327, 0.0285, respectively. experimental results superior those other was effective at bag clouds.

Language: Английский

Citations

0
Determination of Accuracy and Usability of a SLAM Scanner GeoSLAM Zeb Horizon: A Bridge Structure Case Study DOI Creative Commons
R. Urban, Martin Štroner, Jaroslav Braun

et al.

Applied Sciences, Journal Year: 2024, Volume and Issue: 14(12), P. 5258 - 5258

Published: June 18, 2024

The presented paper focuses on testing the performance of a SLAM scanner Zeb Horizon by GeoSLAM for creation digital model bridge construction. A cloud acquired using static Leica ScanStation P40 served as reference. Clouds from both scanners were registered into same coordinate system Trimble S9 HP total station. acquisition was performed independently in two passes. data suffered relatively high noise. Denoising MLS (Moving Least Squares) method to reduce An overall comparison point clouds original and MLS-smoothed data. In addition, ICP (Iterative Closest Point) algorithm also used evaluate local accuracy. RMSDs MLS-denoised approximately 0.02 m Subsequently, more detailed analysis performed, calculating several profiles This revealed that deviations reference did not exceed 0.03 any direction (longitudinal, transverse, elevation) which is, considering length 133 m, very good result. These results demonstrate applicability tested many applications, such twins.

Language: Английский

Citations

3
Airborne LiDAR Point Cloud Classification Using Ensemble Learning for DEM Generation DOI Creative Commons

Ting-Shu Ciou,

Chao‐Hung Lin, Chi‐Kuei Wang

et al.

Sensors, Journal Year: 2024, Volume and Issue: 24(21), P. 6858 - 6858

Published: Oct. 25, 2024

Airborne laser scanning (ALS) point clouds have emerged as a predominant data source for the generation of digital elevation models (DEM) in recent years. Traditionally, DEM using ALS involves steps cloud classification or ground filtering to extract points and labor-intensive post-processing correct misclassified points. The current deep learning techniques leverage ability geometric recognition classification. However, classifiers are generally trained 3D with simple terrains, which decrease performance model inferencing. In this study, point-based boosting ensemble set features inputs is proposed. With strategy, study integrates specialized designed different terrains boost robustness accuracy. experiments, containing various were used evaluate feasibility proposed method. results demonstrated that method can improve quality generated DEMs. accuracy F1 score improved from 80.9% 92.2%, 82.2% 94.2%, respectively, by methods. addition, error, terms mean squared error (RMSE), reduced 0.318-1.362 m 0.273-1.032 learning.

Language: Английский

Citations

3
Deep Ground Filtering of Large-Scale ALS Point Clouds via Iterative Sequential Ground Prediction DOI Creative Commons
Hengming Dai, Xiangyun Hu, Zhen Shu

et al.

Remote Sensing, Journal Year: 2023, Volume and Issue: 15(4), P. 961 - 961

Published: Feb. 9, 2023

Ground filtering (GF) is a fundamental step for airborne laser scanning (ALS) data processing. The advent of deep learning techniques provides new solutions to this problem. Existing deep-learning-based methods utilize segmentation or classification framework extract ground/non-ground points, which suffers from dilemma in keeping high spatial resolution while acquiring rich contextual information when dealing with large-scale ALS due the computing resource limits. To end, we propose SeqGP, novel GF pipeline that explicitly converts task into an iterative sequential ground prediction (SeqGP) problem using points-profiles. proposed SeqGP utilizes reinforcement (DRL) optimize sequence and retrieve bare terrain gradually. 3D sparse convolution integrated strategy generate high-precision results memory efficiency. Extensive experiments on two challenging test sets demonstrate state-of-the-art performance universality method data.

Language: Английский

Citations

7
Scanning the underground: Comparison of the accuracies of SLAM and static laser scanners in a mine tunnel DOI Creative Commons
Martin Štroner,

Rudolf Urban,

Tomáš Křemen

et al.

Measurement, Journal Year: 2024, Volume and Issue: unknown, P. 115875 - 115875

Published: Oct. 1, 2024

Language: Английский

Citations

2
Evaluation of Accuracy in Estimating Diameter at Breast Height Based on the Scanning Conditions of Terrestrial Laser Scanning and Circular Fitting Algorithm DOI Open Access
Yong Kyu Lee, Jung‐Soo Lee

Forests, Journal Year: 2024, Volume and Issue: 15(2), P. 313 - 313

Published: Feb. 7, 2024

A growing societal interest exists in the application of lidar technology to monitor forest resource information and forestry management activities. This study examined possibility estimating diameter at breast height (DBH) two tree species, Pinus koraiensis (PK) Larix kaempferi (LK), by varying number terrestrial laser scanning (TLS) scans (1, 3, 5, 7, 9) DBH estimation methods (circle fitting [CF], ellipse [EF], circle with RANSAC [RCF], [REF]). evaluates combination that yields highest accuracy. The results showed for PK, lowest RMSE 0.97 was achieved when REF applied data from nine after noise removal. For LK, 1.03 observed applying CF seven Furthermore, ANOVA revealed no significant difference estimated more than three were used RCF five EF REF. These are expected be useful establishing efficient accurate plans using TLS monitoring.

Language: Английский

Citations

1