检索结果-内蒙古大学图书馆

Handling data Scarcity Through data Augmentation in Training of deep Neural Networks for 3d data processing

INTERNATIONAL JOURNAL ON SEMANTIC WEB ANd INFORMATION SYSTEMS 2022年第1期18卷 1-16页

作者： Srivastava, Akhilesh Mohan Rotte, Priyanka Ajay Jain, Arushi Prakash, Surya Indian Inst Technol Indore India

due to the availability of cheap 3d sensors such as Kinect and LidAR, the use of 3d data in various domains such as manufacturing, healthcare, and retail to achieve operational safety, improved outcomes, and enhanced customer experience has gained momentum in recent years. In many of these domains, object recognition is being performed using 3d data against the difficulties posed by illumination, pose variation, scaling, etc. present in 2d data. In this work, the authors propose three data augmentation techniques for 3d data in point cloud representation that use sub-sampling. They then verify that the 3d samples created through data augmentation carry the same information by comparing the iterative closest point registration error within the sub-samples, between the subsamples and their parent sample, between the sub-samples with different parents and the same subject, and finally, between the sub-samples of different subjects. They also verify that the augmented sub-samples have the same characteristics and features as those of the original 3d point cloud by applying the central limit theorem.

关键词： 3d Biometrics 3d data processing Computer Vision data Augmentation Object Recognition Sampling

来源：评论

学校读者我要写书评

暂无评论

A novel auricular prosthesis which incorporates a cartilage conduction hearing aid based on 3d data processing technique: a preclinical evaluation

引用

EUROPEAN ARCHIVES OF OTO-RHINO-LARYNGOLOGY 2022年第7期279卷 3741-3744页

作者： Nishiyama, Takanori Hayashi, Shintaro Oishi, Naoki Keio Univ Dept Otolaryngol Head & Neck Surg Sch Med Shinjuku Ku 35 Shinanomachi Tokyo 1608582 Japan Aiwa Prosthet & Orthot Co Ltd Nerima Ku 19-4 Sakaecho Tokyo Japan

Purpose To achieve the challenging goal of simultaneously improving aesthetics and auditory function in the treatment of microtia, we developed an auricular prosthesis incorporating a cartilage conduction hearing aid (APiCHA), which covers a cartilage conduction hearing aid (CC-HA). We evaluated the acoustic characteristics generated by the method of combining these two components using an artificial mastoid and investigated the possibility of clinical translation. Methods The first step in creating an APiCHA is to take molds of both the affected and normal auricles and invert the 3d data of the normal auricle to create data for the auricular prostheses. Grooves were then made inside the prosthetic data to fit the affected ear and a CC-HA, and the APiCHA was made of silicone. The acoustic characteristics were measured using an artificial mastoid, and the results were compared between CC-HA alone and with APiCHA. Results Compared with CC-HA alone, the gain was approximately 2 dB lower at high frequencies from 1 k Hz and higher, and approximately 2 dB higher at approximately 900 Hz when CC-HA was used with APiCHA. For the other frequencies, the acoustic characteristics were almost the same. Conclusion The changes in acoustic characteristics caused by the combined use of APiCHA and CC-HA were minimal and did not pose a clinical problem. The combined use of APiCHA and CC-HA can be considered as a non-invasive and clinically applicable treatment option to achieve both aesthetic and auditory improvements for microtia.

关键词： Atresia Microtia Auricular reconstruction Hearing loss 3d data processing Atresiaplasty

来源：评论

学校读者我要写书评

暂无评论

Producing 3d radargrams from orbital radar sounding data at Mars: History, results, methods, lessons and plans

引用

ICARUS 2024年 419卷

作者： Foss II, Frederick J. Putzig, Nathaniel E. Campbell, Bruce A. Levin, Stewart A. Perry, Matthew R. Holt, John W. Christoffersen, Michael S. Smith, Isaac B. Morgan, Gareth A. Russell, Aaron T. Freestyle Analyt & Quantitat Serv LLC 2210 Parkview Dr Longmont CO 80504 USA Planetary Sci Inst 1700 E Ft Lowell RdSuite 106 Tucson AZ 85719 USA Planetary Sci Inst 405 Urban StSuite 300 Lakewood CO 80228 USA Smithsonian Inst Ctr Earth & Planetary Studies MRC 315POB 37012 Washington DC 20013 USA Stanford Univ Dept Geophys 397 Panama Mall Stanford CA 94305 USA Univ Arizona Lunar & Planetary Lab 1629 E Univ Blvd Tucson AZ 85721 USA York Univ Lassonde Sch Engn 4700 Keele St Toronto ON M3J 1P3 Canada Univ Alaska Fairbanks Geophys Inst 2156 Koyukuk Dr Fairbanks AK 99709 USA

The Shallow Radar (SHARAd) instrument aboard Mars Reconnaissance Orbiter (MRO) has been conducting sounding operations of Mars' near surface for more than 16 years. Results of the SHARAd investigation have been well documented in the literature, with relatively high coverage density in the polar regions leading to some of the earliest scientific discoveries there. data collection and ensuing surprises of SHARAd's primary and first extended science phases were the impetus for choosing the north polar region for producing a first 3d radar image. This early attempt was successful, producing the first 3d radar image of Mars' north polar layered deposits in Planum Boreum, and shortly thereafter a companion 3d radar image of the south polar layered deposits in Planum Australe. Subsequent work refined the 3d methodology used to produce these images, leading to a significantly higher quality 3d image of the former as well as the first 3d radar image of the debris-covered glaciers in the east-central portion of deuteronilus Mensae (dM) in the mid-latitudes. Prior experience in terrestrial seismic data processing and analysis methods forms the basis of the 3d methodology used with Martian radar data, and has been indispensable in ongoing efforts to improve this methodology to further clarify the 3d images of the targets. The purposes of this article are to review 1) the SHARAd 3d work history and results, 2) the methodology developed and challenges encountered thus far in producing the SHARAd 3d images, 3) the broader impact on the processing and analysis of SHARAd data, and 4) current efforts and plans for producing follow-on SHARAd 3d images. Similarities and differences between orbital radar and seismic sounding and data processing are sprinkled throughout the article as reminders that the outcomes in this case are very much the product of cross-disciplinary knowledge and experience.

关键词： Mars 3d orbital radar processing 3d data processing 3d radar imaging Mars polar ices Mars mid-latitude ices

来源：评论

学校读者我要写书评

暂无评论

Application of pseudo-3d sub-bottom profile imaging technology in small submarine target detection

引用

ACTA GEOPHYSICA 2025年第1期73卷 393-405页

作者： Li, Tianguang Huang, Zhiqing Zhang, Xiaobo Meng, Fansheng Pei, Yifan Guo, Jiali Southern Marine Sci & Engn Guangdong Lab Zhuhai Zhuhai 519000 Peoples R China Zhuhai Yunzhou Intelligence Technol LTD Zhuhai 519000 Peoples R China

The sub-bottom profiler is a valuable tool for obtaining high-resolution shallow stratigraphic data in marine geological and geophysical surveys. To detect and acquire the structural characteristics of small submarine objects, we developed a data processing method that utilizes 2d data to construct a 3d structural model. We conducted application experiments using sub-bottom profile detection data from Chuanshan Islands, which were explored using China's most advanced unmanned exploration platform and commercial shallow formation profiling system. To create high-resolution 3d seafloor structure models from recorded 2d sub-bottom profile datasets, an optimized data processing sequence was devised, comprising two stages: 2d data processing and 3d data processing. The 2d data processing stage involved spectrum analysis, band-pass filtering, matching filtering, time-varying gain, and surge correction. The subsequent 3d data processing stage encompassed ping location reallocation, static correction, and extraction of feature layer information. Notably, the final pseudo-3d sub-bottom profile time slice exhibited significant amplitude variations near the target body. This methodology represents an extension of the application of 2d sub-bottom profile data, enhancing the target recognition capabilities of such data. To further improve the precision of target body characterization, we used ArcScene 10.0 to create a 3d sub-bottom profile formation model spatial database. We constructed a submarine 3d formation structure model to show the 3d structural characteristics of the target body in detail and identified a seabed target body measuring 6.4 x 9.2 x 10 m.

关键词： Pseudo-3d sub-bottom profile 2d data processing 3d data processing Target detection

来源：评论

学校读者我要写书评

暂无评论

Measuring the Short-Term degradation of Stone in Historical Buildings

引用

INTERNATIONAL JOURNAL OF ARCHITECTURAL HERITAGE 2024年第10期18卷 1571-1595页

作者： Quintana, Blanca Adan, Antonio Ramon, Amanda UNED Dept Ingn Electr Elect Control Telemat & Quim Apli ETSI Ind Madrid Spain Univ Castilla La Mancha 3D Visual Comp & Robot Lab Ciudad Real Spain UNED Dept Ingn Electr Elect Control Telematica&Quim Apl ETSI Ind CJuan Rosal 12 Madrid 28040 Spain

This paper proposes a methodology with which to measure the slight erosion undergone by valuable heritage pieces in a short period of time. The problems dealt with by our method concern the irregular (free-shape) geometry of the piece to be measured, the difficulty involved in achieving the fine registration required among different samples of the piece, the slight loss of material in a short period of time and the suitable definition of descriptors that will help to carry out a reliable qualitative and quantitative analysis. This document proposes a basic methodology that is able to tackle each of the aforementioned problems. An in-depth explanation of the degradation analysis has been carried out on a representative case study, including the qualitative and quantitative procedures employed. The results obtained lead us to believe that this methodology could be reproduced and applied to similar degradation problems related to old buildings.

关键词： 3d data processing deformation measurement heritage preservation laser scanning point clouds stone degradation assessment

来源：评论

学校读者我要写书评

暂无评论

The 3d GeoInfo Journey from 2006 to 2022: A Reflection

The 3D GeoInfo Journey from 2006 to 2022: A Reflection

引用

18th International 3d GeoInfo Conference

作者： Rahman, Alias Abdul Univ Teknol Malaysia 3D GIS Res Lab Fac Built Environm & Surveying Johor Baharu 81310 Johor Malaysia

ISBN: (纸本)9783031437014;9783031436994;9783031436987

This paper describes research on 3d GIS and the related studies undertaken by the community that were published and presented at the 3d GeoInfo conferences from 2006 to 2022: the first conference was organized in Kuala Lumpur and the 17th conference in Sydney. Several earlier works (published prior to 2006) are also part of the compilation, forming a basis of the 3d GIS research literature and part of the introduction of this paper. Almost all events in the series are described, with the exception of the Wuhan version. The descriptions are categorized into two decades, and the reflections of the 3d GIS research, and development (R-&-d) are highlighted towards the end of the paper.

关键词： 3d spatial data 3d spatial modelling 3d data processing 3d spatial database 3d visualization 3d city modelling digital twins 3d geoinformation

来源：评论

学校读者我要写书评

暂无评论

NeedLR: Streamlining Point Cloud Annotation for Enhanced Machine Learning Integration

NeedLR: Streamlining Point Cloud Annotation for Enhanced Mac...

引用

Practice and Experience in Advanced Research Computing (PEARC)

作者： Stone, Gunner Tavakkoli, Alireza Univ Nevada Reno NV 89557 USA

ISBN: (纸本)9798400704192

With the rising popularity of remote sensing across various scientific and engineering disciplines, the demand for efficient analysis of point cloud data has surged. However, the inherent complexity and volume of point cloud data pose considerable obstacles to human annotation efforts, which are an essential step within the machine learning pipeline for generating accurate training datasets. NeedLR emerges as a solution, offering a robust, userfriendly platform tailored for precise and streamlined point cloud annotation. By harnessing GPU-accelerated visualization, NeedLR facilitates interactive 3d manipulation of point clouds, granting users an intuitive means to dive into their data. Further optimized for efficient RAM usage and employing parallel computing strategies, NeedLR achieves great performance across varied computing environments. Its accessible interface broadens user engagement, rendering it a prime candidate for crowdsourced annotation initiatives and enhancing its utility in machine learning endeavors. This paper presents NeedLR, exploring its development, key features, and the user-centric philosophy that shapes its design. We highlight the potential for NeedLR's role in enhancing current point cloud annotation techniques, merging operational efficiency with broad access to empower users across disciplines.

关键词： point cloud annotation machine learning 3d data processing superpoint-graph complex networks

来源：评论

学校读者我要写书评

暂无评论

Chest area segmentation in 3d images of sleeping patients

引用

MEdICAL & BIOLOGICAL ENGINEERING & COMPUTING 2022年第8期60卷 2159-2172页

作者： Goldstein, Yoav Schatz, Martin Avigal, Mireille Open Univ Israel Dept Math & Comp Sci Raanana Israel Univ Chem & Technol Dept Comp & Control Engn Prague Czech Republic

Although the field of sleep study has greatly developed over recent years, the most common and efficient way to detect sleep issues remains a sleep examination performed in a sleep laboratory. This examination measures several vital signals by polysomnograph during a full night's sleep using multiple sensors connected to the patient's body. Nevertheless, despite being the gold standard, the sensors and the unfamiliar environment's connection inevitably impact the quality of the patient's sleep and the examination itself. Therefore, with the novel development of accurate and affordable 3d sensing devices, new approaches for non-contact sleep study have emerged. These methods utilize different techniques to extract the same breathing parameters but with contactless methods. However, to enable reliable remote extraction, these methods require accurate identification of the basic region of interest (ROI), i.e., the patient's chest area. The lack of automated ROI segmenting of 3d time series is currently holding back the development process. We propose an automatic chest area segmentation algorithm that given a time series of 3d frames containing a sleeping patient as input outputs a segmentation image with the pixels that correspond to the chest area. Beyond significantly speeding up the development process of the non-contact methods, accurate automatic segmentation can enable a more precise feature extraction. In addition, further tests of the algorithm on existing data demonstrate its ability to improve the sensitivity of a prior solution that uses manual ROI selection. The approach is on average 46.9% more sensitive with a maximal improvement of 220% when compared to manual ROI. All mentioned can pave the way for placing non-contact algorithms as leading candidates to replace existing traditional methods used today.

关键词： MS Kinect data acquisition depth sensors Segmentation 3d data processing Human-machine interaction Breathing analysis

来源：评论

学校读者我要写书评

暂无评论

Radiation source location technology of complex space based on point cloud neural network

Radiation source location technology of complex space based ...

引用

2024 International Applied Computational Electromagnetics Society Symposium, ACES-China 2024

作者： Xing, Yu Chen, Guangzhi dong, Aoyong Weng, Youlong Beihang University School of Electronic and Information Engineering China Hefei Innovation Research Institute of Beihang University China

ISBN: (纸本)9798350355581

This paper presents a point cloud neural network (PCNN) to locate radiation sources in complex space. The network has the ability to process 3d sampling information which consists of the electric field intensity and coordinate. The ability of the network to predict the location of the radiation source is obtained by training and fitting the relationship between the sampled electromagnetic information and the radiation source position. Compared with traditional methods, the proposed PCNN method avoids the calculation of electromagnetic wave propagation law in complex space, as well as reduce the computing resources and time required for positioning process. The performance of the proposed method is validated by a series of simulation dataset. © 2024 IEEE.

关键词： 3d data processing point cloud neural network Radiation source location received electric field strength

来源：评论

学校读者我要写书评

暂无评论

Visibility analysis for the occlusion detection and characterisation in street point clouds acquired with Mobile Laser Scanning

引用

GEOCARTO INTERNATIONAL 2022年第25期37卷 10152-10169页

作者： Barros-Ribademar, Jose Balado, Jesus Arias, Pedro Maria Gonzalez-Collazo, Silvia Univ Vigo Escola Enxenaria Ind Vigo Spain Univ Vigo GeoTECH Grp CINTECX Vigo Spain

Although multi-temporal acquisitions are a widely accepted way to reduce occlusions in point clouds, occlusion analysis continues to be a visual analysis. The objective of this work is the design of an automatic method for the detection and characterization of street point cloud occlusions acquired with Mobile Laser Scanning (MLS). The proposed method consists of four main phases: alignment of the point cloud, rasterization, generation of occluded point clouds and visibility analysis. The proposed method was tested with point clouds of TerraMobil-ita/iQmulus dataset. The result shows a correct detection of 98.71% of the occlusions on ground and 83.37% on facades. All detected occlusions were correctly characterized. In the case studies, cars generated the largest amount of occluded area on the ground, with an occluded area per object of 15.36 m2. The main source of occlusions on facades was the building geometry and windows.

关键词： LidAR occlusion recognition occlusion regeneration 3d data processing point cloud completion urban environment

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：