Examinando por Materia "3D reconstruction"
Mostrando1 - 4 de 4
Resultados por página
Opciones de clasificación
- PublicaciónSólo datos3D scene reconstruction Based on a 2D moving LiDAR(Springer, Cham, 2018-10-25) Murcia Moreno, Harold Fabián; Monroy, Maria Fernanda; Mora, Luis FernandoA real-world reconstruction from a computer graphics tool is one of the main issues in two different communities: robotics and artificial intelligence, both of them under different points of view such as computer science, perception and machine vision. A real scene can be reconstructed by generating of point clouds with the help of depth sensors, rotational elements and mathematical transformations according to the mechanical design. This paper presents the development of a three-dimensional laser range finder based on a two-dimensional laser scanner Hokuyo URG-04LX-UG01 and a step motor. The design and kinematic model of the system to generate 3D point clouds are presented with an experimental acquisition algorithm implemented on Robotic Operative System ROS in Python language. The quality of the generated reconstruction is improved with a calibration algorithm based on a model parameter optimization from a reference surface, the results from the calibrated model were compared with a commercial low-cost device. The concurrent application of the system permits the viewing of the scene from different perspectives. The output files can be easily visualized with Python or MATLAB and used for surface reconstruction, scene classification or mapping. In this way, typical robotic tasks can be realized, highlighting autonomous navigation, collision avoidance, grasp calculation and handling of objects.
- PublicaciónSólo datosComparative study of point cloud registration techniques between ICP and others(Proceedings of SPIE - The International Society for Optical Engineering, 2020-08-21) Méndez, Dehyro; Forero, Manuel G.; Murcía, Harold F.Registration is a technique employed for the alignment of point clouds in a single coordinate system. This process is very useful for the reconstruction of 3D plant models, the extraction of their morphological features and the subsequent analysis of the phenotype. One of the most widely studied recording algorithms is ICP (Iterative Closest Point), which is based on rigid transformations. Although in the literature there are several comparative studies between different variants of ICP, there is no comparative study with other more recent existing methods based on other principles. Therefore, in this paper we present a study comparing the results obtained with different registration algorithms on previously filtered 3D point clouds of plants, obtained with a MS Kinect V1 sensor integrated to a rotating base. The study includes two of the most used variants of the ICP, the point-to-point ICP and the point-to-plane ICP. These variants are based on the normals to the surfaces found to guide their point-to-point matching method presenting better results in smooth regions. In addition, other iterative point cloud alignment algorithms based on probability density estimation, hierarchical mixed Gaussian models and distance minimization between probability distributions are included. The results showed the effectiveness of ICP variants simplicity, and the high precision achieved by probabilistic methods. The error and computation time of the algorithms, implemented in Python, were evaluated.
- PublicaciónSólo datosDevelopment of a Low-Cost System for 3D Orchard Mapping Integrating UGV and LiDAR(MDPI - PLANTS, 2021-11-17) Murcia, Harold F.; Tilaguy, Sebastian; Ouazaa, SofianeGrowing evaluation in the early stages of crop development can be critical to eventual yield. Point clouds have been used for this purpose in tasks such as detection, characterization, phenotyping, and prediction on different crops with terrestrial mapping platforms based on laser scanning. 3D model generation requires the use of specialized measurement equipment, which limits access to this technology because of their complex and high cost, both hardware elements and data processing software. An unmanned 3D reconstruction mapping system of orchards or small crops has been developed to support the determination of morphological indices, allowing the individual calculation of the height and radius of the canopy of the trees to monitor plant growth. This paper presents the details on each development stage of a low-cost mapping system which integrates an Unmanned Ground Vehicle UGV and a 2D LiDAR to generate 3D point clouds. The sensing system for the data collection was developed from the design in mechanical, electronic, control, and software layers. The validation test was carried out on a citrus crop section by a comparison of distance and canopy height values obtained from our generated point cloud concerning the reference values obtained with a photogrammetry method. A 3D crop map was generated to provide a graphical view of the density of tree canopies in different sections which led to the determination of individual plant characteristics using a Python-assisted tool. Field evaluation results showed plant individual tree height and crown diameter with a root mean square error of around 30.8 and 45.7 cm between point cloud data and reference values.
- PublicaciónAcceso abiertoLiDAR platform for acquisition of 3d plant phenotyping database(2022-08-25) Comeche, José Manuel; Murcia, Harold F.; Méndez, Dehyro; Martínez Pérez, Juan FranciscoCurrently, there are no free databases of 3D point clouds and images for seedling phenotyping. Therefore, this paper describes a platform for seedling scanning using 3D Lidar with which a database was acquired for use in plant phenotyping research. In total, 362 maize seedlings were recorded using an RGB camera and a SICK LMS4121R-13000 laser scanner with angular resolutions of 45° and 0.5° respectively. The scanned plants are diverse, with seedling captures ranging from less than 10 cm to 40 cm, and ranging from 7 to 24 days after planting in different light conditions in an indoor setting. The point clouds were processed to remove noise and imperfections with a mean absolute precision error of 0.03 cm, synchronized with the images, and time-stamped. The database includes the raw and processed data and manually assigned stem and leaf labels. As an example of a database application, a Random Forest classifier was employed to identify seedling parts based on morphological descriptors, with an accuracy of 89.41%.