Individual Tree Crown Detection Using UAV Orthomosaic

Tahar, Khairul Nizam; Asmadin, M. A.; Sulaiman, Saiful Aman; Khalid, Norlin; Idris, Ahmad Norhisyam; Razali, Muhamad Hasbullah Mohd

doi:10.48084/etasr.4093

Cited by 5 publications

(4 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…TLS involves emitting laser beams to measure distances and create detailed 3D point clouds of surfaces and structures. Photogrammetry, on the other hand, relies on analyzing photographs to extract accurate measurements and generate 3D models [12]. Both methods offer complementary capabilities that enable the acquisition of high-resolution spatial data essential for geospatial mapping, urban planning, environmental monitoring, and infrastructure development.…”

Section: Geospatial Technology Mappingmentioning

confidence: 99%

Transforming Physical Crime Scene into Geospatial-based Point Cloud Data

Azmil,

Mohd Ariff,

Razali

et al. 2024

Eng. Technol. Appl. Sci. Res.

View full text Add to dashboard Cite

Terrestrial Laser Scanning (TLS) and Close-Range Photogrammetry (CRP) are advanced techniques for capturing 3D data in crime scene reconstruction, offering complementary information. Despite taking multiple scans and images from different angles to ensure a comprehensive model, limitations, such as device positioning, shadows, object distance, and laser beam angles prevent the creation of a complete crime scene model. Therefore, combining TLS and CRP data is crucial for achieving a comprehensive reconstruction. This study aims to transform a physical crime scene into a geospatial-based reconstructed model known as point clouds. The technique used was highly rich in realistic features, digitally reconstructed from TLS and CRP. The data sources were then fused via a rigid body transformation, creating a comprehensive crime scene model. The combined point cloud measurements were compared with measurements obtained from a high-precision Vernier caliper to ascertain their accuracy. The resulting Root Mean Square (RMSE) difference between the fused point cloud data and the high-precision caliper measurements was approximately ±4mm. The fusion of TLS and CRP data provides reliable and highly accurate 3D model point clouds, making it suitable for forensic applications.

show abstract

Section: Geospatial Technology Mappingmentioning

confidence: 99%

Transforming Physical Crime Scene into Geospatial-based Point Cloud Data

Azmil,

Mohd Ariff,

Razali

et al. 2024

Eng. Technol. Appl. Sci. Res.

View full text Add to dashboard Cite

show abstract

“…Therefore, it is essential to study the distribution characteristics of tree species 2 – 4 In forest resource management and monitoring, 5 , 6 forest disturbance assessment, 7 and carbon storage calculation, 7 tree species classification and tree statistics are of great significance 8 . Stand density, namely the number of trees per hectare, is an important factor for evaluating site productivity and affecting the growth and development of trees, as well as the main factor affecting the structure and function of forest ecosystems; 9 , 10 it has irreplaceable practical significance for the sustainable development of forestry and the stability of forest ecosystems.…”

Section: Introductionmentioning

confidence: 99%

Multispecies individual tree crown extraction and classification based on BlendMask and high-resolution UAV images

Zhou

Chen

et al. 2023

J. Appl. Rem. Sens.

View full text Add to dashboard Cite

Accurate detection and segmentation of individual trees from unmanned aerial vehicle images is critical for forestry resource surveys and accurate forest management. Deep learning methods have been used for studies of individual tree crown segmentation, classification, and number of trees in mixed coniferous and broad-leaved forests, but the accuracy needs to be improved. Therefore, this study uses BlendMask, a simpler and more efficient algorithm that combines Mask R-CNN and Yolact algorithms to effectively combine instance-level information with semantic information at a finer granularity level, greatly improving crown segmentation accuracy and classification results. Three coniferous species and five broad-leaved species unmanned aerial vehicle images collected from the Jing Yue multispecies ecological forestry site in Changping District, Beijing, were used as the dataset, and the results were compared with Yolact and Mask R-CNN. The results show that the method described in this work has the highest Kappa coefficient (0.89) and overall accuracy (92.14%) in the test set. For segmentation accuracy, coniferous species' producer's accuracy was 0.91 to 0.95, whereas that of broadleaved species was 0.89 to 0.92. For species classification, the F1-score and mean average precision for coniferous species were greater than 91%, whereas those for broad-leaved species were 77.64% to 85.63%. The accuracy of extracting stand density in low and medium canopy density stands was 0.9909 and 0.9422, respectively, whereas that in high canopy density stands was 0.8913. This study shows that the BlendMask model has a good effect in studying the classification of multiple tree species, the segmentation of individual tree crowns, and the statistics of the number of trees in complex forest areas. Compared with broad-leaved forests and high canopy density stands, this model is more suitable for coniferous forest and medium and low canopy density stand scenarios. This study provides an important tool for obtaining more accurate species classification, canopy segmentation, and resource inventory results in complex forest areas.

show abstract

“…The reconstruction of indoor crime scenes can be facilitated by numerous technologies that are beyond photographs and handwritten notes. However, even with modern technologies, it is necessary to take into consideration the technical constraints, user-friendliness, cost, and time duration of data acquisition [11][12]. A recent study reported that leveraging on UAV technology will improve the effectiveness of crime investigation, facilitating access to otherwise inaccessible areas and expediting the retrieval of images [13][14].…”

Section: Introductionmentioning

confidence: 99%

3D Indoor Crime Scene Reconstruction from Micro UAV Photogrammetry Technique

Sazaly,

Mohd Ariff,

Razali

2023

Eng. Technol. Appl. Sci. Res.

View full text Add to dashboard Cite

The application of micro Unmanned Aerial Vehicles (UAVs) in photogrammetry, particularly within the realm of forensic investigation represents a relatively novel approach and has gained increased attention. By measuring the distances and positions of the scene’s components, it is feasible to document and visualize the scene using the photographs that were taken for the purpose of assisting investigators. Capturing accurate crime scene data within a short time frame is always a challenge. Conventionally, photographs were used to document the scene, but the technical qualities of the photographs depended on the skill of the present forensic personnel. The use of 3-Dimensional (3D) photogrammetry enables the production of highly realistic and detailed 3D documentation of a given scene. As this technique involves capturing a series of photographs, it can be a time-consuming process. Therefore, this study aims to explore an alternative approach that enables the rapid acquisition of the scene while preserving the intricate details, thus ensuring efficiency without compromising the accuracy of the resulting documentation. The study employs a methodological approach wherein data are collected from a simulated crime scene situated within a confined and hard-to-reach area. The data collection is facilitated through the utilization of micro UAVs. The acquired data are then processed utilizing photogrammetry software, leading to the generation of a 3D model point cloud. The collected data will be subjected to a comparative analysis with data generated using a Terrestrial Laser Scanner (TLS) as a reference, alongside Vernier Calliper (VC) measurements. The findings indicate that the Root Mean Square Error (RMSE) of the integrated point clouds from TLS and micro UAVs compared to the conventional method is approximately ±0.217 cm. It can be deduced that the integration of data derived from micro UAVs and TLS in forensic photogrammetry within a confined crime scene is viable and yields a high-precision 3D model point cloud.

show abstract

Individual Tree Crown Detection Using UAV Orthomosaic

Cited by 5 publications

References 21 publications

Transforming Physical Crime Scene into Geospatial-based Point Cloud Data

Transforming Physical Crime Scene into Geospatial-based Point Cloud Data

Multispecies individual tree crown extraction and classification based on BlendMask and high-resolution UAV images

3D Indoor Crime Scene Reconstruction from Micro UAV Photogrammetry Technique

Contact Info

Product

Resources

About