To address the challenges of missed detections in water surface target detection using solely visual algorithms in unmanned surface vehicle(USV)perception,this paper proposes a method based on the fusion of visual and...To address the challenges of missed detections in water surface target detection using solely visual algorithms in unmanned surface vehicle(USV)perception,this paper proposes a method based on the fusion of visual and LiDAR point-cloud projection for water surface target detection.Firstly,the visual recognition component employs an improved YOLOv7 algorithmbased on a self-built dataset for the detection of water surface targets.This algorithm modifies the original YOLOv7 architecture to a Slim-Neck structure,addressing the problemof excessive redundant information during feature extraction in the original YOLOv7 network model.Simultaneously,this modification simplifies the computational burden of the detector,reduces inference time,and maintains accuracy.Secondly,to tackle the issue of sample imbalance in the self-built dataset,slide loss function is introduced.Finally,this paper replaces the original Complete Intersection over Union(CIoU)loss function with the Minimum Point Distance Intersection over Union(MPDIoU)loss function in the YOLOv7 algorithm,which accelerates model learning and enhances robustness.To mitigate the problem of missed recognitions caused by complex water surface conditions in purely visual algorithms,this paper further adopts the fusion of LiDAR and camera data,projecting the threedimensional point-cloud data from LiDAR onto a two-dimensional pixel plane.This significantly reduces the rate of missed detections for water surface targets.展开更多
The accuracy and efficiency of three-dimensional(3D)surface forming,which directly affects the cycle and quality of production,is important in manufacturing.In practice,given the uncertainty of metal plate springback,...The accuracy and efficiency of three-dimensional(3D)surface forming,which directly affects the cycle and quality of production,is important in manufacturing.In practice,given the uncertainty of metal plate springback,an error exists between the actual plate and the target surface,which creates a nonlinear mapping from computer aided design models to bending surfaces.Technicians need to reconfigure parameters and process a surface multiple times to delicately control springback,which greatly wastes human and material resources.This study aims to address the springback control problem to improve the efficiency and accuracy of sheet metal forming.A basic computation approach is proposed based on the DeepFit model to calculate the springback value in 3D surface bending.To address the sample data shortage problem,we put forward an advanced approach by combining a deep learning model with case-based reasoning(CBR).Next,a multi-model fused bending parameter generation framework is devised to implement the advanced springback computation approach through surface data preprocessing,CBR-based model matching,convolution neural network-based machining surface generation,and bending parameter generation with a series of model transformations.Moreover,the proposed approaches and the framework are verified by considering saddle surface processing as an example.Overall,this study provides a new idea to improve the accuracy and efficiency of surface processing.展开更多
Point-cloud data acquired using a terrestrial laser scanner play an important role in digital forestry research.Multiple scans are generally used to overcome occlusion effects and obtain complete tree structural infor...Point-cloud data acquired using a terrestrial laser scanner play an important role in digital forestry research.Multiple scans are generally used to overcome occlusion effects and obtain complete tree structural information.However,the placement of artificial reflectors in a forest with complex terrain for marker-based registration is time-consuming and difficult.In this study,an automatic coarse-to-fine method for the registration of pointcloud data from multiple scans of a single tree was proposed.In coarse registration,point clouds produced by each scan are projected onto a spherical surface to generate a series of two-dimensional(2D)images,which are used to estimate the initial positions of multiple scans.Corresponding feature-point pairs are then extracted from these series of 2D images.In fine registration,point-cloud data slicing and fitting methods are used to extract corresponding central stem and branch centers for use as tie points to calculate fine transformation parameters.To evaluate the accuracy of registration results,we propose a model of error evaluation via calculating the distances between center points from corresponding branches in adjacent scans.For accurate evaluation,we conducted experiments on two simulated trees and six real-world trees.Average registration errors of the proposed method were 0.026 m around on simulated tree point clouds,and 0.049 m around on real-world tree point clouds.展开更多
There are diverse products related to human buttocks, which need to be designed, manufactured and evaluated with 3D buttock model. The 3D buttock model used in present research field is just simple approximate model s...There are diverse products related to human buttocks, which need to be designed, manufactured and evaluated with 3D buttock model. The 3D buttock model used in present research field is just simple approximate model similar to human buttocks. The 3D buttock percentile model is highly desired in the ergonomics design and evaluation for these products. So far, there is no research on the percentile sizing system of human 3D buttock model. So the purpose of this paper is to develop a new method for building three-dimensional buttock percentile model in computer system. After scanning the 3D shape of buttocks, the cloud data of 3D points is imported into the reverse engineering software（Geomagic） for the reconstructing of the buttock surface model. Five characteristic dimensions of the buttock are measured through mark-points after models being imported into engineering software CATIA. A series of space points are obtained by the intersecting of the cutting slices and 3D buttock surface model, and then are ordered based on the sequence number of the horizontal and vertical slices. The 1st, 5th, 50 th, 95 th, 99 th percentile values of the five dimensions and the spatial coordinate values of the space points are obtained, and used to reconstruct percentile buttock models. This research proposes a establishing method of percentile sizing system of buttock 3D model based on the percentile values of the ischial tuberosities diameter, the distances from margin to ischial tuberosity and the space coordinates value of coordinate points, for establishing the Nth percentile 3D buttock model and every special buttock types model. The proposed method also serves as a useful guidance for the other 3D percentile models establishment for other part in human body with characteristic points.展开更多
An approach is presented to generate rough interference-free tool-paths directly from massive unorganized data in rough machining that is performed by machining volumes of material in a slice-by-slice manner. Unorgani...An approach is presented to generate rough interference-free tool-paths directly from massive unorganized data in rough machining that is performed by machining volumes of material in a slice-by-slice manner. Unorganized point-cloud is firstly converted to cross-section data. Then a robust data-structure named tool-path net is constructed to save tool-path data. Optimal algorithms for partitioning sub-cut-areas and computing interference-free cutter-locations are put forward. Finally the tool-paths are linked in a zigzag milling mode, which can be transformed into a traveling sales man problem. The experiment indicates optimal tool paths can be acquired, and high computation efficiency can be obtained and interference can be avoided successfully.展开更多
Forest resource management and ecological assessment have been recently supported by emerging technologies.Terrestrial laser scanning(TLS)is one that can be quickly and accurately used to obtain three-dimensional fore...Forest resource management and ecological assessment have been recently supported by emerging technologies.Terrestrial laser scanning(TLS)is one that can be quickly and accurately used to obtain three-dimensional forest information,and create good representations of forest vertical structure.TLS data can be exploited for highly significant tasks,particularly the segmentation and information extraction for individual trees.However,the existing single-tree segmentation methods suffer from low segmentation accuracy and poor robustness,and hence do not lead to satisfactory results for natural forests in complex environments.In this paper,we propose a trunk-growth(TG)method for single-tree point-cloud segmentation,and apply this method to the natural forest scenes of Shangri-La City in Northwest Yunnan,China.First,the point normal vector and its Z-axis component are used as trunk-growth constraints.Then,the points surrounding the trunk are searched to account for regrowth.Finally,the nearest distributed branch and leaf points are used to complete the individual tree segmentation.The results show that the TG method can effectively segment individual trees with an average F-score of 0.96.The proposed method applies to many types of trees with various growth shapes,and can effectively identify shrubs and herbs in complex scenes of natural forests.The promising outcomes of the TG method demonstrate the key advantages of combining plant morphology theory and LiDAR technology for advancing and optimizing forestry systems.展开更多
Self-occlusions are common in rice canopy images and strongly influence the calculation accuracies of panicle traits. Such interference can be largely eliminated if panicles are phenotyped at the 3 D level.Research on...Self-occlusions are common in rice canopy images and strongly influence the calculation accuracies of panicle traits. Such interference can be largely eliminated if panicles are phenotyped at the 3 D level.Research on 3 D panicle phenotyping has been limited. Given that existing 3 D modeling techniques do not focus on specified parts of a target object, an efficient method for panicle modeling of large numbers of rice plants is lacking. This paper presents an automatic and nondestructive method for 3 D panicle modeling. The proposed method integrates shoot rice reconstruction with shape from silhouette, 2 D panicle segmentation with a deep convolutional neural network, and 3 D panicle segmentation with ray tracing and supervoxel clustering. A multiview imaging system was built to acquire image sequences of rice canopies with an efficiency of approximately 4 min per rice plant. The execution time of panicle modeling per rice plant using 90 images was approximately 26 min. The outputs of the algorithm for a single rice plant are a shoot rice model, surface shoot rice model, panicle model, and surface panicle model, all represented by a list of spatial coordinates. The efficiency and performance were evaluated and compared with the classical structure-from-motion algorithm. The results demonstrated that the proposed method is well qualified to recover the 3 D shapes of rice panicles from multiview images and is readily adaptable to rice plants of diverse accessions and growth stages. The proposed algorithm is superior to the structure-from-motion method in terms of texture preservation and computational efficiency. The sample images and implementation of the algorithm are available online. This automatic, cost-efficient, and nondestructive method of 3 D panicle modeling may be applied to high-throughput 3 D phenotyping of large rice populations.展开更多
基金supported by the National Natural Science Foundation of China(No.51876114)the Shanghai Engineering Research Center of Marine Renewable Energy(Grant No.19DZ2254800).
文摘To address the challenges of missed detections in water surface target detection using solely visual algorithms in unmanned surface vehicle(USV)perception,this paper proposes a method based on the fusion of visual and LiDAR point-cloud projection for water surface target detection.Firstly,the visual recognition component employs an improved YOLOv7 algorithmbased on a self-built dataset for the detection of water surface targets.This algorithm modifies the original YOLOv7 architecture to a Slim-Neck structure,addressing the problemof excessive redundant information during feature extraction in the original YOLOv7 network model.Simultaneously,this modification simplifies the computational burden of the detector,reduces inference time,and maintains accuracy.Secondly,to tackle the issue of sample imbalance in the self-built dataset,slide loss function is introduced.Finally,this paper replaces the original Complete Intersection over Union(CIoU)loss function with the Minimum Point Distance Intersection over Union(MPDIoU)loss function in the YOLOv7 algorithm,which accelerates model learning and enhances robustness.To mitigate the problem of missed recognitions caused by complex water surface conditions in purely visual algorithms,this paper further adopts the fusion of LiDAR and camera data,projecting the threedimensional point-cloud data from LiDAR onto a two-dimensional pixel plane.This significantly reduces the rate of missed detections for water surface targets.
基金This work was supported by the National Natural Science Foundation of China(No.61972243).
文摘The accuracy and efficiency of three-dimensional(3D)surface forming,which directly affects the cycle and quality of production,is important in manufacturing.In practice,given the uncertainty of metal plate springback,an error exists between the actual plate and the target surface,which creates a nonlinear mapping from computer aided design models to bending surfaces.Technicians need to reconfigure parameters and process a surface multiple times to delicately control springback,which greatly wastes human and material resources.This study aims to address the springback control problem to improve the efficiency and accuracy of sheet metal forming.A basic computation approach is proposed based on the DeepFit model to calculate the springback value in 3D surface bending.To address the sample data shortage problem,we put forward an advanced approach by combining a deep learning model with case-based reasoning(CBR).Next,a multi-model fused bending parameter generation framework is devised to implement the advanced springback computation approach through surface data preprocessing,CBR-based model matching,convolution neural network-based machining surface generation,and bending parameter generation with a series of model transformations.Moreover,the proposed approaches and the framework are verified by considering saddle surface processing as an example.Overall,this study provides a new idea to improve the accuracy and efficiency of surface processing.
基金funded by the Fundamental Research Funds for the Central Universities(No.2021ZY92)National Students'innovation and entrepreneurship training program(No.201710022076)the State Scholarship Fund from China Scholarship Council(CSC No.201806515050).
文摘Point-cloud data acquired using a terrestrial laser scanner play an important role in digital forestry research.Multiple scans are generally used to overcome occlusion effects and obtain complete tree structural information.However,the placement of artificial reflectors in a forest with complex terrain for marker-based registration is time-consuming and difficult.In this study,an automatic coarse-to-fine method for the registration of pointcloud data from multiple scans of a single tree was proposed.In coarse registration,point clouds produced by each scan are projected onto a spherical surface to generate a series of two-dimensional(2D)images,which are used to estimate the initial positions of multiple scans.Corresponding feature-point pairs are then extracted from these series of 2D images.In fine registration,point-cloud data slicing and fitting methods are used to extract corresponding central stem and branch centers for use as tie points to calculate fine transformation parameters.To evaluate the accuracy of registration results,we propose a model of error evaluation via calculating the distances between center points from corresponding branches in adjacent scans.For accurate evaluation,we conducted experiments on two simulated trees and six real-world trees.Average registration errors of the proposed method were 0.026 m around on simulated tree point clouds,and 0.049 m around on real-world tree point clouds.
文摘There are diverse products related to human buttocks, which need to be designed, manufactured and evaluated with 3D buttock model. The 3D buttock model used in present research field is just simple approximate model similar to human buttocks. The 3D buttock percentile model is highly desired in the ergonomics design and evaluation for these products. So far, there is no research on the percentile sizing system of human 3D buttock model. So the purpose of this paper is to develop a new method for building three-dimensional buttock percentile model in computer system. After scanning the 3D shape of buttocks, the cloud data of 3D points is imported into the reverse engineering software（Geomagic） for the reconstructing of the buttock surface model. Five characteristic dimensions of the buttock are measured through mark-points after models being imported into engineering software CATIA. A series of space points are obtained by the intersecting of the cutting slices and 3D buttock surface model, and then are ordered based on the sequence number of the horizontal and vertical slices. The 1st, 5th, 50 th, 95 th, 99 th percentile values of the five dimensions and the spatial coordinate values of the space points are obtained, and used to reconstruct percentile buttock models. This research proposes a establishing method of percentile sizing system of buttock 3D model based on the percentile values of the ischial tuberosities diameter, the distances from margin to ischial tuberosity and the space coordinates value of coordinate points, for establishing the Nth percentile 3D buttock model and every special buttock types model. The proposed method also serves as a useful guidance for the other 3D percentile models establishment for other part in human body with characteristic points.
文摘An approach is presented to generate rough interference-free tool-paths directly from massive unorganized data in rough machining that is performed by machining volumes of material in a slice-by-slice manner. Unorganized point-cloud is firstly converted to cross-section data. Then a robust data-structure named tool-path net is constructed to save tool-path data. Optimal algorithms for partitioning sub-cut-areas and computing interference-free cutter-locations are put forward. Finally the tool-paths are linked in a zigzag milling mode, which can be transformed into a traveling sales man problem. The experiment indicates optimal tool paths can be acquired, and high computation efficiency can be obtained and interference can be avoided successfully.
基金The work was supported by the National Natural Science Foundation of China(Grant Number 41961060)the Key Program of Basic Research of Yunnan Province,China(Grant Number 2019FA017)+1 种基金the Multi-government International Science and Technology Innovation Cooperation Key Project of National Key Research and Development Program of China(Grant Number 2018YFE0184300)the Program for Innovative Research Team in Science and Technology research and innovation fund(ysdyjs 2020058)in the University of Yunnan Province.
文摘Forest resource management and ecological assessment have been recently supported by emerging technologies.Terrestrial laser scanning(TLS)is one that can be quickly and accurately used to obtain three-dimensional forest information,and create good representations of forest vertical structure.TLS data can be exploited for highly significant tasks,particularly the segmentation and information extraction for individual trees.However,the existing single-tree segmentation methods suffer from low segmentation accuracy and poor robustness,and hence do not lead to satisfactory results for natural forests in complex environments.In this paper,we propose a trunk-growth(TG)method for single-tree point-cloud segmentation,and apply this method to the natural forest scenes of Shangri-La City in Northwest Yunnan,China.First,the point normal vector and its Z-axis component are used as trunk-growth constraints.Then,the points surrounding the trunk are searched to account for regrowth.Finally,the nearest distributed branch and leaf points are used to complete the individual tree segmentation.The results show that the TG method can effectively segment individual trees with an average F-score of 0.96.The proposed method applies to many types of trees with various growth shapes,and can effectively identify shrubs and herbs in complex scenes of natural forests.The promising outcomes of the TG method demonstrate the key advantages of combining plant morphology theory and LiDAR technology for advancing and optimizing forestry systems.
基金supported by the National Natural Science Foundation of China (U21A20205)Key Projects of Natural Science Foundation of Hubei Province (2021CFA059)+1 种基金Fundamental Research Funds for the Central Universities (2021ZKPY006)cooperative funding between Huazhong Agricultural University and Shenzhen Institute of Agricultural Genomics (SZYJY2021005,SZYJY2021007)。
文摘Self-occlusions are common in rice canopy images and strongly influence the calculation accuracies of panicle traits. Such interference can be largely eliminated if panicles are phenotyped at the 3 D level.Research on 3 D panicle phenotyping has been limited. Given that existing 3 D modeling techniques do not focus on specified parts of a target object, an efficient method for panicle modeling of large numbers of rice plants is lacking. This paper presents an automatic and nondestructive method for 3 D panicle modeling. The proposed method integrates shoot rice reconstruction with shape from silhouette, 2 D panicle segmentation with a deep convolutional neural network, and 3 D panicle segmentation with ray tracing and supervoxel clustering. A multiview imaging system was built to acquire image sequences of rice canopies with an efficiency of approximately 4 min per rice plant. The execution time of panicle modeling per rice plant using 90 images was approximately 26 min. The outputs of the algorithm for a single rice plant are a shoot rice model, surface shoot rice model, panicle model, and surface panicle model, all represented by a list of spatial coordinates. The efficiency and performance were evaluated and compared with the classical structure-from-motion algorithm. The results demonstrated that the proposed method is well qualified to recover the 3 D shapes of rice panicles from multiview images and is readily adaptable to rice plants of diverse accessions and growth stages. The proposed algorithm is superior to the structure-from-motion method in terms of texture preservation and computational efficiency. The sample images and implementation of the algorithm are available online. This automatic, cost-efficient, and nondestructive method of 3 D panicle modeling may be applied to high-throughput 3 D phenotyping of large rice populations.
