Three-dimensional Reconstruction Method Research Articles

Map-based localization for intelligent vehicles based on fusion of multiple visual features in underground parking

The ‘last mile’ problem in intelligent car localization within underground parking lots presents challenges such as localization blind spots and feature extraction difficulties. To tackle these issues, this paper proposes a mapping and multiscale matching localization method based on the fusion of multiple feature sources using a ‘binocular mapping, monocular localization’ approach. By combining traditional features with deep learning features, the proposed method enhances feature matching accuracy and addresses feature extraction challenges. A sparse three-dimensional reconstruction method based on local feature-optical flow is also proposed to mitigate localization blind spots. During the mapping process, the underground parking lot scene is segmented into several nodes to store map information, each consisting of a feature layer, scene structure layer, and trajectory layer. In the multiscale matching localization process, ORB, LoFTR, and BoW multisource features are fused to achieve feature matching, determine the nearest node, and estimate vehicle pose using the PnP model through optical flow. Finally, a localization optimization algorithm outputs the localization result. Experiments were conducted across multiple underground parking lots encompassing a total area of over 100,000 square meters. The results demonstrated that the average error of map construction using the proposed algorithm is approximately 15 mm, and the average error rate of localization is 4.45%.

Influence of Solute Drag Effect and Interphase Precipitation of Nb on Ferrite Transformation.

The significant impact of Nb on ferrite transformation, both in terms of solute drag effect (SDE) and interphase precipitation, was investigated quantitatively. Ferrite transformation kinetics were characterized using thermal expansion experiments and theoretical calculations. The microstructures were characterized using high-temperature confocal laser scanning microscopy (CLSM), a field-emission scanning electron microscope (FESEM), and a transmission electron microscope (TEM). Under a higher driving force, interphase precipitations were observed in the sample with a higher Nb content. A three-dimensional (3D) reconstruction method was used to convert the two-dimensional (2D) image of interphase precipitation into a three-dimensional model for a more typical view. The SDE and interphase precipitation had opposite effects on the kinetics of ferrite transformation. A lower Nb content showed a strong contribution to the SDE, which delayed ferrite transformation. A higher concentration of Nb was expected to enhance the SDE, but the inhibition effect was eliminated by the interphase precipitation of NbC during interfacial migration. Both the experimental results and theoretical calculations confirmed this phenomenon.

An effective point cloud registration method for three-dimensional reconstruction of pressure piping

Abstract At present, industrial scenes with sparse features and weak textures are widely encountered, and the three-dimensional reconstruction of such scenes is a recognized problem. Pressure pipelines have a wide range of applications in fields such as petroleum engineering, chemical engineering, and hydropower station engineering. However, there is no mature solution for the three-dimensional reconstruction of pressure pipes. The main reason is that the typical scenes in which pressure pipes are found also have relatively few features and textures. Traditional three-dimensional reconstruction algorithms based on feature extraction are largely ineffective for such scenes that are lacking in features. In view of the above problems, this paper proposes an improved interframe registration algorithm based on point cloud fitting with cylinder axis vector constraints. By incorporating geometric feature parameters of a cylindrical pressure pipeline, specifically the axis vector of the cylinder, to constrain the traditional iterative closest point algorithm, the accuracy of point cloud registration can be improved in scenarios lacking features and textures, and some environmental uncertainties can be overcome. Finally, using actual laser point cloud data collected from pressure pipelines, the proposed fitting-based point cloud registration algorithm with cylinder axis vector constraints is tested. The experimental results show that under the same conditions, compared with other open-source point cloud registration algorithms, the proposed method can achieve higher registration accuracy. Moreover, integrating this algorithm into an open-source three-dimensional reconstruction algorithm framework can lead to better reconstruction results.

Influence of the Road Model on the Optimal Maneuver of a Racing Motorcycle

Motorcycle motion is largely influenced by the road geometry, which alters the allowable accelerations in longitudinal and lateral directions and influences the vertical wheel loads. Recently, a method for three-dimensional road reconstruction and its incorporation into transient and quasi-steady-state (QSS) minimum lap time simulations (MLTSs) has been proposed. The main purpose of this work is to demonstrate how significantly different results from a minimum lap time optimal control problem can be obtained when using inappropriate elevation data sources in the track reconstruction problem, and how the road model reconstructed using poor input data can lead to misleading conclusions when analyzing real vehicle and driver performances. Two road models derived from high- and low-resolution digital elevation models (DEMs) are compared and their impact on the optimal maneuver of a racing motorcycle is examined. The essentials of track identification are presented, as well as vehicle positioning on the 3D road and the generalized QSS motorcycle model. Obtained 3D and 2D road models are analyzed in detail, on a case example of the Road Atlanta racetrack, and used in minimum lap time simulations, which are validated by the experimental data recorded on the Supersport motorcycle. The comparative analysis showed that great care should be taken when selecting the elevation dataset in the track reconstruction process, and that the 1 m resolution local DEMs seem to be sufficient to obtain MLTS results close to the measured ones. The example of using the 3D free-trajectory QSS minimum lap time problem to localize the track segments where real driver actions can be improved is also presented. The differences between simulation results on different road models of the same racetrack can be large and influence the interpretation of optimal maneuver.

A color 3D reconstruction method for waste mechanical parts based on improved SGM

In order to improve the remanufacturing efficiency of scrap mechanical parts and comprehensively detect their surface fault status, this paper proposes a color three-dimensional reconstruction method of scrap mechanical parts based on an improved semi-global matching (SGM) algorithm. In experiments, this method demonstrated significant performance advantages in dealing with complex mechanical component structures and large illumination interference environments. Experimental results show that the three-dimensional color model reconstructed by this method has clear texture and small dimensional error, and is suitable for online analysis of surface fault information of scrap mechanical parts in actual production lines. Through quantitative analysis, compared with the traditional SGM method, the method in this paper improves the structural similarity index (SSIM) by an average of 19.8% and reduces the mean square error (MSE) by an average of 33.1%.

Three-dimensional reconstruction method of rail corrugation based on the vision detection

In order to realise the visual identification of the three-dimensional geometric characteristics of the rail corrugation, the three-dimensional reconstruction method of rail corrugation based on vision detection is developed in the present paper. The acquisition equipment is firstly designed to acquire the images of rail corrugation and corresponding depth information. Then, the image preprocessing method is proposed to achieve the automatic segmentation of rail corrugation images. The fitting equation between the image and depth of rail corrugation is obtained, which allows the three-dimensional reconstruction of rail corrugation. Furthermore, based on the finite element model of the wheel-rail system and the wear calculation model, the practicality and effectiveness of the three-dimensional reconstruction method are verified. Results show that the depth-grayscale correlation equation is well fitted with the coefficient of determination of 0.94 and the root mean square error of 0.02. The three-dimensional reconstruction method has an error of 2.63% at the maximum wear depth. When the three-dimensional reconstruction method is applied to the wear prediction of rail corrugation, the wear prediction of rail corrugation with the three-dimensional reconstruction is approximately the same as the actual wear distribution, which has an error of 5.02% at the maximum wear depth. The above results demonstrate the effectiveness of the three-dimensional reconstruction method of rail corrugation based on the vision detection.

Human-Following Strategy for Orchard Mobile Robot Based on the KCF-YOLO Algorithm

Autonomous mobile robots play a vital role in the mechanized production of orchards, where human-following is a crucial collaborative function. In unstructured orchard environments, obstacles often obscure the path, and personnel may overlap, leading to significant disruptions to human-following. This paper introduces the KCF-YOLO fusion visual tracking method to ensure stable tracking in interference environments. The YOLO algorithm provides the main framework, and the KCF algorithm intervenes in assistant tracking. A three-dimensional binocular-vision reconstruction method was used to acquire personnel positions, achieving stabilized visual tracking in disturbed environments. The robot was guided by fitting the personnel’s trajectory using an unscented Kalman filter algorithm. The experimental results show that, with 30 trials in multi-person scenarios, the average tracking success rate is 96.66%, with an average frame rate of 8 FPS. Additionally, the mobile robot is capable of maintaining a stable following speed with the target individuals. Across three human-following experiments, the horizontal offset Error Y does not exceed 1.03 m. The proposed KCF-YOLO tracking method significantly bolsters the stability and robustness of the mobile robot for human-following in intricate orchard scenarios, offering an effective solution for tracking tasks.

Establishment and preliminary application of personalized three-dimensional reconstruction of thyroid gland with automatic detection of thyroid nodules based on ultrasound videos.

A well display of the spatial location of thyroid nodules in the thyroid is important for surgical path planning and surgeon-patient communication. The aim of this study was to establish a three-dimensional (3D) reconstruction method of the thyroid gland, thyroid nodule, and carotid artery with automatic detection based on two-dimensional (2D) ultrasound videos, and to evaluate its clinical value. Ultrasound videos, including the thyroid gland with nodule, isthmus of thyroid gland, and ipsilateral carotid artery, were recorded. BC-UNet, MTN-Net, and RDPA-U-Net network models were innovatively employed for segmentation of the thyroid glands, the thyroid nodules, and the carotid artery respectively. Marching Cubes algorithm was used for reconstruction, while Laplacian smoothing algorithm was employed to smooth the 3D model surface. Using this model, 20 patients and 15 surgeons completed surveys on the effectiveness of this model for the pre-surgery demonstration of nodule location as well as surgeon-patient communication. The thyroid gland with nodule, isthmus of gland, and carotid artery were reconstructed and displayed. With the 3D model, the understanding of the spatial location of thyroid nodules improved in all three surgeon groups, eliminating the influence of professional levels. In the patient survey, the patients' understanding of the thyroid nodule location and procedure for surgery were significantly improved. In addition, with the 3D model, the time for doctors to explain to patients was significantly reduced (16.75 vs. 8.85min, p=0.001). To our knowledge, this is the first report of constructing a 3D thyroid model using a deep learning technique for personalized thyroid segmentation based on 2D ultrasound videos. The preliminary clinical application showed that it was conducive to the comprehension of the location of thyroid nodules for surgeons and patients, with significant improvement on the efficiency of surgeon-patient communication.

Clinical and economic analysis of the application of the method of three-dimensional reconstruction of lung tissue in the practice of tuberculosis institution

Background. The use of clinical and economic analysis in practical healthcare allows for a more rational use of the resources allocated by the state and contributes to the achievement of optimal indicators in terms of the quality of disease identification and the effectiveness of its treatment.
 Aim. To conduct a clinical and economic analysis of the prospects for introducing the 3D lung reconstruction method into tuberculosis practice, which is used to correct the level of adherence, reduce the severity of symptoms of anxiety and depression, and increase the effectiveness of chemotherapy.
 Materials and methods. In this scientific work, the results of treatment (more than 90 people) of newly diagnosed patients with destructive pulmonary tuberculosis, which were observed in the Nizhny Novgorod Regional Clinical Tuberculosis Dispensary, were used. The patients were divided into two groups. For the 1st group of patients, tactile lung models were made, among them a questionnaire was conducted using the MMAS-4 questionnaire and the HADS scale. For patients of the 2nd group, lung models were not made, and no survey was conducted. Each group was assessed the effectiveness of chemotherapy. Clinical and economic analysis of the cost of introducing the method of three-dimensional lung reconstruction into tuberculosis practice was carried out on the basis of two evaluation criteria «cost – effectiveness» and «cost-effectiveness increment» with additional mathematical calculations of the necessary indicators. The results obtained were compared with the existing known method of increasing adherence to tuberculosis treatment - the issuance of food packages.
 Results. Based on the results of the study, it was proved that the use of tactile models of the lungs in phthisiatric practice is more than 8 times cheaper than the method of issuing food kits with, in fact, comparable results of treatment effectiveness. If there is a need to increase the effectiveness of the treatment of tuberculosis patients by increasing the subsidy of food packages, the cost of one additional percent of efficiency is not economically viable.
 Conclusion. The clinical and economic analysis of the prospects for introducing the method of three-dimensional lung reconstruction into phthisiatric practice presented in the paper looks promising, which is confirmed by the performed mathematical calculations and comparisons, and can be recommended for use in the routine practice of phthisiatricians to correct the level of adherence of tuberculosis patients to chemotherapy, reduce the severity symptoms of anxiety and depression, as well as an increase in the effectiveness of treatment.

Bi-level Analysis of Computed Tomography Images of Malignant Pleural Mesothelioma: Deep Learning-Based Classification and Subsequent Three-Dimensional Reconstruction

Abstract A bi-level analysis of computed tomography (CT) images of malignant pleural mesothelioma (MPM) is presented in this paper, starting with a deep learning-based system for classification, followed by a three-dimensional (3D) reconstruction method. MPM is a highly aggressive cancer caused by asbestos exposure, and accurate diagnosis and determination of the tumor’s volume are crucial for effective treatment. The proposed system employs a bi-level approach, utilizing machine learning and deep learning techniques to classify CT lung images and subsequently calculate the tumor’s volume. The study addresses challenges related to deep neural networks, such as the requirement for large and diverse datasets, hyperparameter optimization, and potential data bias. To evaluate performance, two convolutional neural network (CNN) architectures, Inception-v3 and ResNet-50, were compared in terms of their features and performance. Based on CT images, the second stage incorporates 3D volume reconstruction. The process is carried out by cropping, registering, filtering, and segmenting images. This study demonstrated the efficacy of the developed system by combining CNN optimizations with 3D image reconstruction. It is intended to improve the accuracy of MPM diagnosis and to assist in the determination of chemotherapy doses, both of which may result in improved outcomes for patients.

Emergency neurosurgical care for patients with acute cerebral circulation disorders

Aim. To improve the results of neurosurgical treatment of patients with acute cerebrovascular disorders.Material and Methods. An analysis of the results of neurosurgical treatment of 48 patients admitted to the Department of Neurosurgery of the State Institution National Medical Center - “Shifobakhsh” was carried out. The age of patients ranged from 30 to 75 years. In each case, patients underwent clinical and laboratory studies, including using CT and MRI, as well as using three-dimensional (3D) and four-dimensional (4D) reconstruction methods of cerebral vessels.Result and Discussion. The most common cause leading to the formation of intracranial hematomas was hypertension, found in 32 (66.6%) patients, while other causes were identified in 12 (25%) patients (including angiopathy, cerebral vascular pathologies, amyloidosis, and the use of certain medications, among others). All patients underwent neurosurgical operations. The highest frequency of recurrences (4.1%) was observed after surgical interventions performed within the first 24 hours following the hemorrhagic event.Conclusion. With early detection of intracranial hematomas and with the optimal choice of method for their treatment, good functional results are observed with a low incidence of deaths.

A Study on the Effects of Photogrammetry by the Camera Angle of View Using Computer Simulation

During the decommissioning activities, a movie was shot inside the reactor building during the investigation of the primary containment vessel by applying photogrammetry, which is one of the methods for three-dimensional (3D) reconstruction from images, to the images from this movie, it is feasible to perform 3D reconstruction of the environment around the primary containment vessel. However, the images from this movie may not be suitable for 3D reconstruction because they were shot remotely by robots owing to limited illumination, high-dose environments, etc. Moreover, photogrammetry has the disadvantage of easily changing 3D reconstruction results by simply changing the shooting conditions. Therefore, this study investigated the accuracy of the 3D reconstruction results obtained by photogrammetry with changes in the camera angle of view under shooting conditions. In particular, we adopted 3D computer graphics software to simulate shooting target objects for 3D reconstruction in a dark environment while illuminating them with light for application in decommissioning activities. The experimental results obtained by applying artificial images generated by simulation to the photogrammetry method showed that more accurate 3D reconstruction results can be obtained when the camera angle of view is neither too wide nor too narrow when the target objects are shot and surrounded. However, the results showed that the accuracy of the obtained results is low during linear trajectory shooting when the camera angle of view is wide.

Feature Extraction Algorithm of Food Spoilage Early Warning Data Based on Machine Learning and Internet of Things

Abstract In view of the problems of the current method, such as the large error, low detection efficiency, and poor analysis ability of food deterioration early warning, a spectral feature extraction algorithm of food spoilage early warning data based on machine learning and Internet of Things was proposed. Based on Internet of Things technology, collect food spoilage warning images. The k-means algorithm is used to identify the food variable mass spectrometry data, and the three-dimensional visual feature reconstruction method is devised to reconstruct the image characteristics of the food variable mass spectrum, and the spectral characteristics of the food deterioration warning data are obtained. Based on machine learning, the features of food variable mass spectrometry image are decomposed, and the gray pheromone of food variable mass spectrometry image is extracted. Under the convolutional neural network structure, the spectral features of food deterioration early warning data are vectorized. Based on the results of spectral vectorization, the spectral features of food deterioration early warning data are extracted. The experimental results show that the proposed method can effectively improve the efficiency of food deterioration detection and reduce the error of food deterioration detection for safe environment.

Are Artificial Intelligence-Assisted Three-Dimensional Histological Reconstructions Reliable for the Assessment of Trabecular Microarchitecture?

Objectives: This study aimed to create a three-dimensional histological reconstruction through the AI-assisted classification of tissues and the alignment of serial sections. The secondary aim was to evaluate if the novel technique for histological reconstruction accurately replicated the trabecular microarchitecture of bone. This was performed by conducting micromorphometric measurements on the reconstruction and comparing the results obtained with those of microCT reconstructions. Methods: A bone biopsy sample was harvested upon re-entry following sinus floor augmentation. Following microCT scanning and histological processing, a modified version of the U-Net architecture was trained to categorize tissues on the sections. Detector-free local feature matching with transformers was used to create the histological reconstruction. The micromorphometric parameters were calculated using Bruker's CTAn software (version 1.18.8.0, Bruker, Kontich, Belgium) for both histological and microCT datasets. Results: Correlation coefficients calculated between the micromorphometric parameters measured on the microCT and histological reconstruction suggest a strong linear relationship between the two with p-values of 0.777, 0.717, 0.705, 0.666, and 0.687 for BV/TV, BS/TV, Tb.Pf Tb.Th, and Tb.Sp, respectively. Bland-Altman and mountain plots suggest good agreement between BV/TV measurements on the two reconstruction methods. Conclusions: This novel method for three-dimensional histological reconstruction provides researchers with a tool that enables the assessment of accurate trabecular microarchitecture and histological information simultaneously.

Three-dimensional reconstruction optimization of tunnel face and intelligent extraction of discontinuity orientation based on binocular stereo vision

In the process of grading and dynamically optimizing the design and construction parameters of the surrounding rock mass of a rock tunnel face, efficiently and accurately acquiring the geometrical parameters of the rock discontinuities is an important basic task. To address the problems of time consuming, low accuracy, and high danger associated with traditional methods of obtaining the structural information of rock mass, this paper proposes a method for three-dimensional reconstruction and intelligent information extraction of tunnel face based on binocular stereo vision (BSV). First, the parallel binocular device with a single camera was improved, calibrated using the checkerboard calibration method. By integrating with the semi-global matching algorithm, the BSV based method for the three-dimensional reconstruction of the rock mass of the tunnel face was optimized. Furthermore, based on the results from on-site engineering applications, this study leveraged two parameters, point cloud density and algorithm runtime, to determine the optimal values for the disparity range and window size parameters within the semi-global stereo matching algorithm. This enhancement improved the performance of the 3D reconstruction method based on binocular stereo vision. Finally, efficient and refined intelligent methods for extracting structural parameters of the rock mass were proposed based on k-nearest neighbor search and kernel density estimation. The research results can provide reliable technical support for the intelligent and efficient acquisition of rock mass structural information in rock tunnel engineering faces.

Monitoring the filling of the landfill using the results of aerial photography

The important task of landfill current residual capacity assessment was solved in this work. The modern methods of high-precision three-dimensional reconstruction were used by the results of unmanned aerial vehicle survey, equipped with a global navigation satellite system receiver. The main stages of work were combining the project and field survey data into one coordinate and height system, reconstruction of designed underground and above-ground parts of landfill, project and actual model comparison, the calculation. As a result, the estimation of the landfill residual capacity becomes possible, which allows to develop recommendations for further landfill operation. The proposed approach allows rapid and high-quality active monitoring of the engineered facility. Photogrammetric processing of the results of low-altitude aerial photography makes it possible to obtain objective data on the current actual state of the landfills, to carry out competent and valid management of the landfill functioning. Such studies can significantly extend the landfill’s lifetime, minimize the negative impact on the environment and predict the attainment of projected capacity much more accurately.

IMetaTown: A Metaverse System with Multiple Interactive Functions Based on Virtual Reality.

This work aims to pioneer the development of a real-time interactive and immersive Metaverse Human-Computer Interaction (HCI) system leveraging Virtual Reality (VR). The system incorporates a three-dimensional (3D) face reconstruction method, grounded in weakly supervised learning, to enhance player-player interactions within the Metaverse. The proposed method, two-dimensional (2D) face images, are effectively employed in a 2D Self-Supervised Learning (2DASL) approach, significantly optimizing 3D model learning outcomes and improving the quality of 3D face alignment in HCI systems. The work outlines the functional modules of the system, encompassing user interactions such as hugs and handshakes and communication through voice and text via blockchain. Solutions for managing multiple simultaneous online users are presented. Performance evaluation of the HCI system in a 3D reconstruction scene indicates that the 2DASL face reconstruction method achieves noteworthy results, enhancing the system's interaction capabilities by aiding 3D face modeling through 2D face images. The experimental system achieves a maximum processing speed of 18 frames of image data on a personal computer, meeting real-time processing requirements. User feedback regarding social acceptance, action interaction usability, emotions, and satisfaction with the VR interactive system reveals consistently high scores. The designed VR HCI system exhibits outstanding performance across diverse applications.

基于时空分析的手术场景三维重建方法研究

基于时空分析的手术场景三维重建方法研究

Three-Dimensional Structure Reconstruction System of Nasal Cavity Based on a Short-Source Acoustic Tube

The nasal cavity three-dimensional structure reconstruction is significant for the diagnosis and treatment of nasal and nasopharynx diseases. Currently, most systems for nasal cavity three-dimensional structure reconstruction are based on a long-source acoustic tube. However, the long-source acoustic tube has the weaknesses of low portability and high cost. To address these issues, a nasal cavity three-dimensional structure reconstruction system based on a short-source acoustic tube was designed in this research. The designed system comprises the lower computer of a nasal acoustic signal acquisition device and the upper computer of nasal acoustic signal analysis software. The lower computer of a nasal acoustic signal acquisition device consists of a signal processing unit, a detection unit, a microcontroller, and a USB communication interface. It is used to collect the original acoustic signal in the nasal cavity. The upper computer of nasal acoustic signal analysis software consists of the nasal cavity acoustic signal preprocessing method and the nasal cavity three-dimensional structure hierarchical reconstruction method. The nasal cavity acoustic signal preprocessing method is used to eliminate the DC offset component and suppress the high-frequency noise. Then, the nasal cavity three-dimensional structure hierarchical reconstruction method is used to establish the relationship between the cross-sectional area of the nasal cavity and the depth of the nasal cavity. In order to verify the effectiveness of the system designed in this research, the straight tube model and the simulated nasal cavity model were selected as test subjects, and the depth–cross-sectional area was selected as the evaluation indicator of the accuracy of the nasal cavity three-dimensional structure reconstruction result. The experimental results show that the root mean squared error and the mean relative error of the depth–cross-sectional area were reduced from 0.6284 cm2 to 0.0201 cm2 and 47.6467% to 1.8248%, respectively, under the straight tube model by using the nasal cavity acoustic signal preprocessing method. The root mean squared error and the mean relative error of the depth–cross-sectional area were reduced from 4.7730 cm2 to 0.1510 cm2 and 107.8128% to 3.3096%, respectively, under the simulated nasal cavity model. Meanwhile, this system can increase the effective measurement depth of the nasal cavity from 7 cm to 13 cm based on the preprocessing method. The results demonstrated that the designed system can not only reconstruct and visualize a nasal cavity three-dimensional structure with high precision but also generate comprehensive test reports. Therefore, this system can provide a basis for further auxiliary diagnosis of the disease and facilitate doctors in explaining the physiological structure of the nasal cavity to patients.

A low-cost regularization strategy for the 3D reconstruction of wind turbines based on remote sensing images

ABSTRACT Unmanned aerial vehicle remote sensing images are easy to acquire and cost-effective. Their application provides a new approach for the maintenance of offshore wind farms. To obtain an accurate model of wind turbines, thereby improving detection efficiency and mitigating the risks associated with high-altitude workings, a low-cost method for wind turbine three-dimensional (3D) reconstruction is proposed. The main advantages of deep learning-based Multi-View Stereo Network (MVSNet) include high accuracy, fast processing speed, and ease of extension to multiple viewpoints. Most networks utilize multi-scale 3D Convolutional Neural Network (CNN) for regularization. However, when dealing with images that have high depth and resolution, CNN regularization strategy leads to a significant increase in memory usage. In this paper, the two-dimensional cost maps are regularized along the depth dimension sequentially using Stacked Convolutional Gated Recurrent Unit (Stacked Conv-GRU), instead of reducing the 3D cost volume all at once. Localized convolution operations replace undifferentiated fully connected operations, substantially reducing memory consumption and enabling high-resolution reconstruction. We conduct comparisons and model evaluations on the ”Nordtank” wind turbine remote sensing image dataset. Experimental results demonstrate that, compared to other 3D reconstruction methods, Stacked Conv-GRU regularized MVSNet reduces run time and hardware requirements while maintaining similar accuracy.