3D Reconstruction Technology Research Articles

Research on Anatomical Types and Variations of Lung Segment Bronchi Based on Artificial Intelligence 3D Reconstruction Technology

Aims: To investigate the anatomic type and variation of segmental bronchus based on artificial intelligence 3D reconstruction technology. Methods: The image data of 200 patients who underwent thin-layer enhanced chest CT examination in the Department of Thoracic Surgery of Shandong Public Health Clinical Center from December 2021 to December 2022 were selected. The bronchi and blood vessels of the affected side of the lung were reconstructed according to the CT scanning data of the patients, and the anatomic types and anatomic variations of the bronchi of each lung segment on the affected side were observed and statistically analyzed. Results: There were basically four anatomical types of the right superior lobar bronchus, of which the three branching subtypes B1, B2 and B3 are the most common anatomical types. There are two types of right middle lobe bronchus, of which the B4 and B5 subtypes are the most common anatomic types. For the left lung, 4 types of segmental bronchus were found and (B1+2a+b, B1+2c), (B3a, B3b+c) was the most common anatomical type. Two types of bronchus were found in the left superior lobe of the lung. Type B4 and type B5 were the most common anatomical types. In addition, there were 3 types of bronchi in the dorsal segment of the left inferior lobe, and 3 types in the anterior, external and posterior basal segment bronchi of the left inferior lobe, among which B8, B9+B10 (57.6%) were the most common. Conclusion: The anatomical types of segmental and subsegmental bronchus are complex and varied, and there may be rare anatomical variations. Therefore, before performing relevant operations, it is necessary to be familiar with the relevant operations and the branch types of corresponding bronchus in advance to ensure the smooth operation.

A Comprehensive Review of Vision-Based 3D Reconstruction Methods.

With the rapid development of 3D reconstruction, especially the emergence of algorithms such as NeRF and 3DGS, 3D reconstruction has become a popular research topic in recent years. 3D reconstruction technology provides crucial support for training extensive computer vision models and advancing the development of general artificial intelligence. With the development of deep learning and GPU technology, the demand for high-precision and high-efficiency 3D reconstruction information is increasing, especially in the fields of unmanned systems, human-computer interaction, virtual reality, and medicine. The rapid development of 3D reconstruction is becoming inevitable. This survey categorizes the various methods and technologies used in 3D reconstruction. It explores and classifies them based on three aspects: traditional static, dynamic, and machine learning. Furthermore, it compares and discusses these methods. At the end of the survey, which includes a detailed analysis of the trends and challenges in 3D reconstruction development, we aim to provide a comprehensive introduction for individuals who are currently engaged in or planning to conduct research on 3D reconstruction. Our goal is to help them gain a comprehensive understanding of the relevant knowledge related to 3D reconstruction.

Development of a 3D point cloud reconstruction-based apple canopy liquid sedimentation model

China leads the world in apple production, contributing to over half of the global total. Annually, extensive pesticide use is employed in apple cultivation, but the uncertain quantities of pesticide deposition and indiscriminate application have resulted in significant pesticide loss and residue. This poses a considerable potential risk to human health. This paper focuses on ground-based plant protection and stratifies the canopy horizontally, using 3D point cloud reconstruction technology and tree deposition experiments to explore the relationship between canopy leaf area index and pesticide deposition. By calculating the droplet penetration coefficient based on the amount of solution deposited at various canopy layers, a functional model linking leaf area index and droplet penetration coefficient has been established, with a correlation coefficient (R2) of 0.92. Based on this model, and by integrating the canopy volume with the amount of solution deposited on the outermost layer of the canopy, the total solution deposition on the entire tree can be estimated. According to the deposition model, there is a negative correlation between the amount of solution deposited on the canopy and the leaf area index of the canopy. Moreover, the experiments have shown that when the spray vehicle travels at a speed of 1.5 m/s, the maximum amount of solution is deposited while still ensuring an effective deposition rate. The establishment of the deposition estimation model allows for the visualization of canopy solution deposition and can guide fruit farmers in determining the appropriate frequency of spraying based on the severity of pest and disease infestations.

Large-Scale 3D Reconstruction from Multi-View Imagery: A Comprehensive Review

Three-dimensional reconstruction is a key technology employed to represent virtual reality in the real world, which is valuable in computer vision. Large-scale 3D models have broad application prospects in the fields of smart cities, navigation, virtual tourism, disaster warning, and search-and-rescue missions. Unfortunately, most image-based studies currently prioritize the speed and accuracy of 3D reconstruction in indoor scenes. While there are some studies that address large-scale scenes, there has been a lack of systematic comprehensive efforts to bring together the advancements made in the field of 3D reconstruction in large-scale scenes. Hence, this paper presents a comprehensive overview of a 3D reconstruction technique that utilizes multi-view imagery from large-scale scenes. In this article, a comprehensive summary and analysis of vision-based 3D reconstruction technology for large-scale scenes are presented. The 3D reconstruction algorithms are extensively categorized into traditional and learning-based methods. Furthermore, these methods can be categorized based on whether the sensor actively illuminates objects with light sources, resulting in two categories: active and passive methods. Two active methods, namely, structured light and laser scanning, are briefly introduced. The focus then shifts to structure from motion (SfM), stereo matching, and multi-view stereo (MVS), encompassing both traditional and learning-based approaches. Additionally, a novel approach of neural-radiance-field-based 3D reconstruction is introduced. The workflow and improvements in large-scale scenes are elaborated upon. Subsequently, some well-known datasets and evaluation metrics for various 3D reconstruction tasks are introduced. Lastly, a summary of the challenges encountered in the application of 3D reconstruction technology in large-scale outdoor scenes is provided, along with predictions for future trends in development.

Research on Three-dimensional Reconstruction

Nowadays, with the continuous development of the field of computer vision and computer graphics, 3D reconstruction technology has become an area that has attracted much of people’s attention. 3D reconstruction refers to creating a 3D model of an object in the real world by extracting information from multiple 2D images or sensor data. This technology has important usages in many fields, including virtual reality, augmented reality, cultural heritage preservation, medical image processing, industrial design, etc. This thesis examines the application of 3D reconstruction in various fields, discusses the definition of development history, its wide application in life, and its advantages and disadvantages. This article is divided into four parts and discusses the algorithm research of 3D reconstruction in four fields. Through an in-depth study of 3D reconstruction technology, we can better understand its role in different fields and provide strong support for developing and innovating related applications. As more accurate devices and algorithms continue to advance, 3D reconstruction technology is expected to show its significant potential in more and more fields and create more realistic and useful digital experiences for us.

Three-dimensional printed personalised digital guide plate for greater palatine block in trigeminal neuralgia

The nerve block is a safe and effective method to theat trigeminal neuralgia (TN). In terms of the V2 trigeminal neuralgia, the most difficult procedure in nerve block is accurate and fast greater palatine foramen (GPF) insertion. In this study, we developed a new technique using a personalised digital tooth-supported guide plate to increase insertion accuracy and success rates and reduce the pain of patients during injection. A total of 18 patients with TN (11 female and 7 male) were enrolled and treated between September 2020 and June 2022. Before injection, the guide plate was designed via Mimics three-dimensional (3D) reconstruction technology and printed via 3D printer. Then, all patients underwent maxillary nerve block with a guide plate for each injection. In this study, placement of all guide plates was completed within one minute and all punctures were successful the first time. The depth of the injection needle was over 2.5 cm in all cases and the guide plate was stability-supported by the maxillary teeth. The various pain scores had an obvious improvement. No patients presented symptoms of local anaesthetic toxicity or onset of new neurological sequelae. Using this new technology, we can significantly reduce the difficulty of GPF insertion and decrease patient pain during injection. The enhanced success rate of nerve block can achieve better therapeutic effect. For surgeons, personalised digital tooth-supported guide plates make the operation easier, especially for novice surgeons.

A method for calculating and simulating phenotype of soybean based on 3D reconstruction

In plant phenotypic research, accurate organ segmentation and crop simulation are crucial for optimizing crop planting and increasing yield. In this study, an efficient method for soybean organ segmentation and phenotypic growth simulation was explored based on 3D reconstruction technology. Taking Dongnong252 soybean as the research object, the soybean phenotype acquisition system based on Kinect sensor was used to achieve high-precision and non-destructive acquisition of soybean canopy images during the whole growth period. First, conditional filtering and statistical filtering were used to remove noise. Then, combined with the Intrinsic Shape Signatures-Coherent Point Drift (ISS-CPD) and the Iterative Closest Point (ICP) algorithms, the multi-view three-dimensional (3D) canopy structure of soybean was reconstructed, which provided a reliable basis for plant stem and leaf segmentation. On this basis, the average accuracy of plant stem and leaf segmentation was 79.99% by using the Distance-field-based segmentation pipeline (DFSP) algorithm. Furthermore, the accurate information was provided for extracting and calculating 3D phenotypic parameters such as plant height, crown width, stem thickness, leaf length and leaf width from three scales of whole plant, stem and leaf. The calculated values were highly consistent with the measured values, with an average coefficient of determination of 0.9654 and an average percentage error of 3.4862%. Meanwhile, by analyzing the quantitative relationship between phenotypic parameters and physiological development time, the data-driven Richards growth simulation model was introduced to accurately predict the growth process of soybean plants. The coefficient of determination R2 values of each phenotypic simulation model reached above 0.9357, which improved the goodness of fit of the model by 0.03 compared with the Logistic model, and its root mean square error (RMSE) ranged from 0.0020 to 0.1112. The research results indicated that this method had high accuracy and reliability in 3D reconstruction of soybean canopy, phenotype calculation, and growth simulation. It could provide quantitative indicators for soybean variety selection, planting, and management, and provide technical support and reference for high-throughput phenotype acquisition and analysis of field crops.

An ultra-fast and energy-efficient CNTFET-based image corner detection hardware for real-time image processing applications

In image processing and machine vision, corner detection is pivotal in diverse applications, including computer vision, 3D reconstruction, face detection, object tracking, and video technologies. Despite the wide usage, the real-time and energy-efficient hardware implementation of corner detection algorithms remains a critical challenge because of the computational resource limitations. On the other hand, owing to the complicated nature of corner detection algorithms, their hardware implementation has been limited to the graphics processing unit (GPU) and field programmable gate arrays (FPGA) platforms. In this regard, this work aims to propose a novel and ultra-efficient carbon nanotube field-effect transistor (CNTFET)-based hardware for image corner detection. Thanks to the proposed corner detection algorithm, the designed hardware has been realized using 2742 transistors with competitive accuracy. The proposed corner detection hardware indicated a remarkable salt-and-pepper noise immunity without using any noise reduction circuit. Our comprehensive simulations demonstrate 78%, 87%, and 94.5% total average improvements in delay, power, and energy compared to the other related corner detection hardware. Moreover, the proposed CNTFET-based corner detection hardware shows a 43 ps propagation delay, demonstrating its real-time operation. The proposed corner detection algorithm at the system level shows suitable accuracy metrics such as Recall, Precision, and error of detection (EoD) compared to the other well-known corner detectors. Our method has established a new pathway for real-time circuit-level hardware design for image processing and machine vision applications.

Application of 3D reconstruction and 3D printing technology in advanced ovarian cancer surgery: a retrospective study.

Advanced ovarian cancer is frequently accompanied by extensive peritoneal metastasis, complicating surgical interventions. This study aims to explore the application of 3D reconstruction and 3D printing technology in the treatment of advanced ovarian cancer. We conducted a retrospective analysis of 60 patients with stage III ovarian cancer who underwent cytoreductive surgery at Hebei University Affiliated Hospital between 2020 and 2023. Patients were randomly assigned to three groups: a 3D visualization group, a 3D visualization plus 3D printing group, and a traditional 2D CT imaging evaluation group. High-precision medical imaging techniques (e.g., CT, MRI) were employed to create digital 3D models, which were then converted into physical entities using 3D printing for surgical planning and simulation. Both the 3D visualization group and the 3D visualization plus 3D printing group demonstrated superior outcomes in terms of surgery duration and blood loss compared to the traditional 2D CT group, indicating the efficacy of 3D reconstruction and 3D printing in preoperative planning. Postoperative recovery indicators, such as hospital stay and time to first flatus, were also more favorable in the groups utilizing 3D technology. Although there were no significant differences in postoperative complications and recurrence rates among the three groups, the groups using 3D technology showed advantages in reducing certain complications. The results indicate that medical 3D technology has significant value in the surgical planning of advanced ovarian cancer, enhancing surgical precision and reducing intraoperative risks, which may aid in improving postoperative recovery.

Overview of image-based 3D reconstruction technology

Three-dimensional (3D) reconstruction technology is the key technology to establish and express the objective world by using computer, and it is widely used in real 3D, automatic driving, aerospace, navigation and industrial robot applications. According to different principles, it is mainly divided into methods based on traditional multi-view geometry and methods based on deep learning. This paper introduces the above methods from the perspective of three-dimensional space representation. The feature extraction and stereo matching theory of traditional 3D reconstruction methods are the theoretical basis of 3D reconstruction methods based on deep learning, so the paper focuses on them. With the development of traditional 3D reconstruction methods and the development of deep learning related theories, the explicit deep learning 3D reconstruction method represented by MVSNet and the implicit 3D reconstruction method represented by NeRF have been gradually developed. At the same time, the dataset and evaluation indicators for 3D reconstruction were introduced. Finally, a summary of image based 3D reconstruction was provided.

3D face imaging and reconstruction technology： a review

3D face imaging and reconstruction technology： a review

Research on Conservation and Restoration Methods of Museum Artifacts in the Context of Artificial Intelligence

Abstract In the era of artificial intelligence, the conservation and restoration of museum artifacts encounter both new possibilities and hurdles. The limitations of conventional conservation techniques, in terms of efficiency and accuracy, are becoming increasingly apparent. This study explores the potential of 3D reconstruction technologies, SIFT algorithms, and Poisson’s equation to revolutionize the digital restoration of cultural relics. These advanced approaches allow for precise, efficient restorations and the ability to undertake complex projects without risking harm to the original items. Our approach integrates cutting-edge algorithms to tailor 3D reconstruction strategies for museum artifacts, culminating in creating a comprehensive digital artifact database. Through the innovative use of feature point detection and texture synthesis, we have achieved notable successes, including a 12% improvement in 3D model accuracy and a significant enhancement in the automation of the restoration process. With an SSIM value of 0.9964 for restored images, our method demonstrates superiority over traditional restoration techniques, marking a significant stride towards the efficient digital preservation of cultural heritage.

Single-photon 3D reconstruction technology of rotating objects

Single-photon 3D reconstruction technology of rotating objects

Экспериментальная оценка рекомендаций метода по сопровождению оператора во время съемки фотографий для последующей 3D-реконструкции

The availability of 3D reconstruction technologies expands the possibilities of their application for analyzing the state of complex technical objects. However, solutions for 3D reconstruction still heavily depend on the quality of the input data and require significant time investments to build high-quality 3D models. The goal of this study is to expedite the process of constructing detailed and defect-free 3D models by creating an automated method to assist the operator during the capture of RGB images for subsequent 3D reconstruction. The method is used for preliminary analysis of the collected data and allows for the identification of problematic frames in advance, as well as determining the areas of the object to be captured that will be reconstructed with defects or low level of detail. The study describes algorithms for the two final stages of the method - global analysis and recommendations, based on the real-time construction of sparse point clouds and 3D models, as well as their properties assessment in conjunction with the operator's trajectory, reconstructed from accelerometer and gyroscope data. The effectiveness of the final stages of the method, based on the proposed algorithms, was evaluated through experiments. It was demonstrated that the use of global analysis and recommendation stages not only allows for predicting defective areas and low-detail zones in dense 3D models on mobile devices, but also accelerates the process of obtaining a high-quality 3D model by at least five times through automated analysis of the collected data.

Corrosion damages of reinforced concrete characterized by X-ray CT and DVC techniques

X-ray microcomputed tomography (X-ray CT) technique was performed to monitor the whole corrosion cracking process of a cylinder reinforced concrete specimen non-destructively. Based on gray threshold separation and 3D reconstruction technology, the spatial distributions of each substance were obtained. Combined with Digital Volume Correlation (DVC) technique, the development of corrosion depth, generation and migration of corrosion products, variation of internal strain field and crack propagation during the corrosion process were all analyzed. The experiment results indicate that the electric current accelerated method can lead to a non-uniformity corrosion at the early age. The distribution fields of rusts and strain are affected by the micro- and meso- structure of concrete. A turning point of maximum principle strain along the cracking path can be identified as the critical point at which micro-cracks of concrete connect and propagate into macro-cracks. The statistical analysis reveals that the gray values of mortar matrix, corrosion products, steel bar and coarse aggregates in corroded concrete follow a Garrison distribution, while the pores do not.

Customised 3D-Printed Surgical Guide for Breast-Conserving Surgery after Neoadjuvant Chemotherapy and Its Clinical Application.

For patients eligible to undergo breast-conserving surgery (BCS) after neoadjuvant chemotherapy, accurate preoperative localisation of tumours is vital to ensure adequate tumour resection that can reduce recurrence probability effectively. For this reason, we have developed a 3D-printed personalised breast surgery guide (BSG) assisted with supine magnetic resonance imaging (MRI) and image 3D reconstruction technology, capable of mapping the tumour area identified on MRI onto the breast directly using dual positioning based on the manubrium and nipple. In addition, the BSG allows the colour dye to be injected into the breast to mark the tumour region to be removed, yielding more accurate intraoperative resection and satisfactory cosmetic outcomes. The device has been applied to 14 patients from January 2018 to July 2023, with two positive margins revealed by the intraoperative biopsy. This study showed that the BSG-based method could facilitate precise tumour resection of BCS by accurately localising tumour extent and margin, promoting the clinical efficacy in patients with breast cancer as well as simplifying the surgical process.

Stereo matching method based on high-precision occlusion-recovering and discontinuity-preserving

Stereo matching is the core of machine vision, such as 3D reconstruction, posture perception and driverless technology in the future etc. Currently, there are still great difficulties in improving the matching accuracy of disparity discontinuity and occlusion regions. In this paper, we propose a new high-precision occlusion-recovering and discontinuity-preserving method for stereo matching. We focus on disparity refinement, and present an adaptability-based disparity reconstruction that achieves accurate occlusion-recovering by fusing color aberration, distance, and disparity value. For discontinuity-preserving, disparity-based secondary guided filtering is proposed, then introduce adaptive image enhancement and adaptive Canny edge detection to correct errors in discontinuous regions. The experiments on Middlebury standard dataset show that our algorithm has high overall accuracy, especially in the disparity discontinuity and occlusion regions. This novel method is more direct, efficient, and less sensitive to the accuracy of disparity estimation, and the live-action shooting shows the robustness of our algorithm. The method has the prospect of civil and engineering use, and presents scientific reference for high-precision target identification, target tracking, 3D measurement, etc.

A Study of 3D model voxelization method for artificial intelligence learning

<p class="abstract">The 3D reconstruction technology which is one of various restoration technologies, implements and shapes 2D pixels of an object that actually exists in a 3D form. As an accurate 3D model for artificial intelligence learning information pre-processing is required. The pre-processing needs to accurately size information and coordinate information of a 3D object for artificial intelligence learning. Also, the 3D model data generated during the preprocessing can be represented as a point cloud, which is point-based coordinate data. In this study, the pixel data was analyzed by voxelizing the 3D model for artificial intelligence learning, and the 3D model data was digitized to form a mesh file. In this study, 3D modeling was done with a 3D modeling tool and the object was exported to STL. In addition, it was converted into mesh file data including 3D coordinate data in Python language on Google’s colab platform. The mesh data created in this way is used in DataSet for artificial intelligence learning (CNN, RNN, GAN). Currently, there are many datasets for 2D artificial intelligence learning, but this study provided an opportunity to collect 3D artificial intelligence learning datasets. The research results can be used in the fields of robots, autonomous driving, games, and product design.</p>

Research on the application of virtual reality technology in contemporary environmental design

Abstract The development and application of virtual reality technology can help the innovative application of environmental design. This paper first focuses on the application of virtual reality technology in the natural environment based on the development of environmental design and the integration of natural geographical conditions. In fact, the focus is on the analysis of virtual reality technology. Selected 3D reconstruction technology algorithms, including camera calibration profile, pair of polar geometry and single response matrix, matching cost and quality assessment index. Finally, the natural environment simulation experiment is constructed based on virtual reality technology. The average real-time rendering FPS of the unoptimized vegetation distribution interval decreases continuously with the increase in the number of vegetation models. When the average vegetation number is around 54,672, the average FPS decreases to around per 53. When the average vegetation count is at 97,713, the average FPS drops to about 28 frames per second. The average real-time dye FPS stabilizes at around 50 after using the LOD multi-detail hierarchical model.

Remaining Useful Life Prediction Method for High Temperature Blades of Gas Turbines Based on 3D Reconstruction and Machine Learning Techniques

Turbine blades are crucial components exposed to harsh conditions, such as high temperatures, high pressures, and high rotational speeds. It is of great significance to accurately predict the life of blades for reducing maintenance cost and improving the reliability of gas turbine systems. A rapid and accurate blade life assessment method holds significant importance in the maintenance plan of gas turbine engines. In this paper, a novel on-line remaining useful life (RUL) prediction method for high-temperature blades is proposed based on 3D reconstruction technology and data-driven surrogate mode. Firstly, the 3D reconstruction technology was employed to establish the geometric model of real turbine blades, and the fluid–thermal–solid analysis under actual operational conditions was carried out in ANSYS software. Six checkpoints were selected to estimate the RUL according to the stress–strain distribution of the blade surface. The maximum equivalent stress was 1481.51 MPa and the highest temperature was 1393.42 K. Moreover, the fatigue-creep lifetime was calculated according to the parameters of the selected checkpoints. The RUL error between the simulation model and commercial software (Control and Engine Health Management (CEHM)) was less than 0.986%. Secondly, different data-driven surrogate models (BP, DNN, and LSTM algorithms) were developed according to the results from numerical simulation. The maximum relative errors of BP, DNN, and LSTM models were 0.030%, 0.019%, and 0.014%. LSTM demonstrated the best performance in predicting the RUL of turbine blades with time-series characteristics. Finally, the LSTM model was utilized for predicting the RUL within a gas turbine real operational process that involved five start–stop cycles.