Structure From Motion Research Articles

Area of origin estimation from multiple arbitrarily oriented surfaces using marker-guided structure from motion

Bloodstain pattern analysis plays a crucial role in forensic investigations. Projected patterns can offer valuable insights into the dynamics of crime scenes. In this paper, we propose and validate a novel approach that extends existing software, HemoVision, to analyze impact patterns that are distributed across multiple arbitrarily oriented surfaces. The proposed method integrates HemoVision’s marker-based system with structure from motion (SfM) techniques to reconstruct the three-dimensional geometry of impact patterns using only two-dimensional photographs. Controlled experiments were used to validate the proposed approach, demonstrating robustness in reconstruction accuracy with median translation errors below 3 mm and median angular errors below 0.2°, irrespective of imaging device or image resolution. Comparing the estimated areas origin to their known ground truth, the proposed method achieved an average total error of 8.12 cm, with the primary source of error being the vertical dimension. Despite this, the overall error remains well within the ranges of error reported in prior work. This study demonstrates that HemoVision can be used to analyze complex impact patterns using only two-dimensional photographs, providing forensic experts with an efficient and accessible tool for investigating intricate crime scenes involving multi-surface impact patterns.

Semihierarchical reconstruction and weak‐area revisiting for robotic visual seafloor mapping

AbstractDespite impressive results achieved by many on‐land visual mapping algorithms in the recent decades, transferring these methods from land to the deep sea remains a challenge due to harsh environmental conditions. Images captured by autonomous underwater vehicles, equipped with high‐resolution cameras and artificial illumination systems, often suffer from heterogeneous illumination and quality degradation caused by attenuation and scattering, on top of refraction of light rays. These challenges often result in the failure of on‐land Simultaneous Localization and Mapping (SLAM) approaches when applied underwater or cause Structure‐from‐Motion (SfM) approaches to exhibit drifting or omit challenging images. Consequently, this leads to gaps, jumps, or weakly reconstructed areas. In this work, we present a navigation‐aided hierarchical reconstruction approach to facilitate the automated robotic three‐dimensional reconstruction of hectares of seafloor. Our hierarchical approach combines the advantages of SLAM and global SfM that are much more efficient than incremental SfM, while ensuring the completeness and consistency of the global map. This is achieved through identifying and revisiting problematic or weakly reconstructed areas, avoiding to omit images and making better use of limited dive time. The proposed system has been extensively tested and evaluated during several research cruises, demonstrating its robustness and practicality in real‐world conditions.

Large dump critical zone of interest identification and slope failure prediction using realistic 3D numerical modelling

Dump failure is one of the most catastrophic hazards in the opencast mining industry and involves sudden and violent failures. Therefore, susceptible region identification is essential for slope failure mitigation in the large dump. This study proposed a sustainable approach for large dump critical zone of interest (CZI) identification, stability monitoring and slope failure prediction. UAV close-range photogrammetry survey and structure-from-motion (SfM) approach were used for the dump realistic 3D model reconstruction. 3D numerical modelling was employed for dump stability analysis. In this investigation, the dump critical zone, slope factor of safety (FOS), displacement and slope failure prediction were analysed using the 3D numerical modelling. The results indicate that maximum displacement was found in the critical zone I. However, the dump was stable under static conditions. Dump critical slope failure time is predicted using the inverse velocity method. The results of this investigation are reliable and establish the scope of UAV photogrammetry for realistic 3D modelling. In addition, the slope stability state and failure model were consistent with field observations. The proposed approach can be used for the mine, hill, and roadcut slopes. UAV technology and 3D modelling approaches have substantially reduced data collection time and expense.

Digital Fracture Porosity Measurement using Digital Outcrop Modelling (DOM)

Abstract Fracture measurements for porosity calculations often experience problems such as barely accessible locations, limited area coverage, and a long time to carry out field data measurements. Therefore, refined alternative methods are necessary to obtain the most favourable data. This study demonstrates fracture measurements through application of Digital Outcrop Modelling (DOM) using the Structure from Motion (SfM) method to utilize digital photogrammetry techniques. This method was chosen because data collection can be done remotely using Light Detection and Ranging (LiDAR) or digital photogrammetry to obtain point cloud data that is used to perform three-dimensional outcrop modeling. In addition, SfM also produces high-resolution data without the need for experts, is cost-efficient and safer than other methods. Hand-held platform Canon 700D digital single lens reflex (DSLR) camera was used in this study to capture data from an area of a square meter located on the riverbed that requires detailed modeling. DOM using SfM is applied to conduct aperture measurements in fractures inside the scan window, later used as data for calculating basement fracture porosity predictions. Field data collection is also carried out to validate data generated by DOM. Calculations achieved from field data and DOM data will be compared to determine differences of the porosity prediction values. Field data collection was carried out on 3 scan windows, each one is an area of 1 m2 and in totals obtained 117 fracture data, consist of 44 extension joints, 34 release joints, 7 shear joints, and 32 non-systematic joints. In this study, 115 photos were taken from scanwindow 1, 162 photos from scanwindow 2, and 158 photos from scanwindow 3 to construct the DOM using agisoft metashape software. The difference in these values can be used as a correction formula or evaluation for further research. The results of measuring basement fracture porosity using data generated by DOM will remain in the range of less than 1%.

A Case Study of Integrating Terrestrial Laser Scanning (TLS) and Building Information Modeling (BIM) in Heritage Bridge Documentation: The Edmund Pettus Bridge

The Edmund Pettus Bridge, Selma, Alabama, a symbol of the American Civil Rights Movement and an exemplar of early 20th-century engineering, stands as a testament to the progress and challenges of its era. The bridge, recognized for its pivotal role in the 1965 “Bloody Sunday” conflict and the following Selma to Montgomery marches for voting rights, also represents significant engineering achievements with its distinctive design and construction methodology. In this study, the research team presents a comprehensive framework for documenting heritage bridges by utilizing Terrestrial Laser Scanning (TLS) technology, supplemented by other Reality Capture (RC) techniques, including Structure from Motion (SfM), 360-degree photography, and Unmanned Aerial Vehicle (UAV), and integrating the data within a Building Information Modeling (BIM) environment. The focus on the Edmund Pettus Bridge case study demonstrates how this novel approach can capture the intricate details of its structural and architectural features while preserving its historical narratives. The documentation outcomes, including a detailed BIM model and a set of Historic American Engineering Record (HAER) drawings, highlight the effectiveness of combining TLS and BIM in conserving unconventional heritage structures like bridges. This paper also discusses the technological challenges encountered, such as dealing with heavy traffic and environmental constraints during data acquisition and developing the BIM model and drawings. It outlines the strategies implemented to address these issues. This research contributes to preserving a severely under-represented American National Historic Landmark (NHL). It sets a precedent for documenting other non-building heritage structures, balancing technological advancements with historical integrity.

Improving 3D reconstruction for accurate measurement of appearance characteristics in shiny fruits using post-harvest particle film: A case study on tomatoes

Besides the importance of internal quality, the textural quality properties of the product have a remarkable effect on consumer satisfaction and market price. Mainly, the three-dimensional (3D) modeling technique is applied to evaluate the quality of agricultural products. This study aimed to employ the structure from motion (SFM) method along with the particle film technique to predict the physical features of tomatoes. Reconstructing 3D models of tomatoes is challenging due to their shiny skin. In this study, a new method based on particle films was developed to address this issue. Fifty samples of tomatoes were randomly selected with various sizes and four ripeness levels with no-external damage. The SFM technique applied with kaolin particle film was a suitable approach for assessing the surface area, volume, and color index, with a correlation coefficient of 0.76, 0.88, and 0.67, respectively. Also, 3D model predictions achieved these textural properties by obtaining RMSEs of 2.44, 2.30, and 0.35, respectively. Finally, the proposed SFM method’s results showed an error of less than 5%. This technique has proven to be a cable technique for the 3D reconstruction of products with a glossy surface.

Development of large-scale synthetic 3D point cloud datasets for vision-based bridge structural condition assessment

Visual recognition of 3D point cloud data of bridge inspection scenes is a key step in automating the visual inspection process, which is currently largely manual and inefficient. To alleviate the lack of large-scale annotated point cloud datasets for training such 3D visual recognition algorithms, this research investigates an approach for developing large-scale synthetic point cloud datasets. The proposed approach proceeds in four steps: (1) random generation of different types of bridges in computer graphics environments; (2) sampling of camera trajectories that represent data collection scenarios during bridge inspection; (3) 3D reconstruction using Structure from Motion (SfM) applied to rendered synthetic images; (4) automated annotation of the reconstructed point cloud using ground truth masks obtained with synthetic images. Besides, this research proposes to store point uncertainty information defined by the error between the ground truth depth and the depth calculated from the SfM results. Prior to training, thresholds can be applied to this uncertainty information to control the levels of outliers in the dataset. This research demonstrates the proposed approach by generating point cloud datasets for two data collection scenarios. The effectiveness of the generated datasets is investigated by training 3D semantic segmentation algorithms and evaluating the performance on real and synthetic point cloud data. The proposed approach for point cloud dataset generation will facilitate the development of generalizable and high level-of-detail 3D recognition algorithms toward autonomous bridge inspection.

Effects of different ground segmentation methods on the accuracy of UAV-based canopy volume measurements.

The nonuniform distribution of fruit tree canopies in space poses a challenge for precision management. In recent years, with the development of Structure from Motion (SFM) technology, unmanned aerial vehicle (UAV) remote sensing has been widely used to measure canopy features in orchards to balance efficiency and accuracy. A pipeline of canopy volume measurement based on UAV remote sensing was developed, in which RGB and digital surface model (DSM) orthophotos were constructed from captured RGB images, and then the canopy was segmented using U-Net, OTSU, and RANSAC methods, and the volume was calculated. The accuracy of the segmentation and the canopy volume measurement were compared. The results show that the U-Net trained with RGB and DSM achieves the best accuracy in the segmentation task, with mean intersection of concatenation (MIoU) of 84.75% and mean pixel accuracy (MPA) of 92.58%. However, in the canopy volume estimation task, the U-Net trained with DSM only achieved the best accuracy with Root mean square error (RMSE) of 0.410 m3, relative root mean square error (rRMSE) of 6.40%, and mean absolute percentage error (MAPE) of 4.74%. The deep learning-based segmentation method achieved higher accuracy in both the segmentation task and the canopy volume measurement task. For canopy volumes up to 7.50 m3, OTSU and RANSAC achieve an RMSE of 0.521 m3 and 0.580 m3, respectively. Therefore, in the case of manually labeled datasets, the use of U-Net to segment the canopy region can achieve higher accuracy of canopy volume measurement. If it is difficult to cover the cost of data labeling, ground segmentation using partitioned OTSU can yield more accurate canopy volumes than RANSAC.

Machine Learning Vegetation Filtering of Coastal Cliff and Bluff Point Clouds

Coastal cliffs erode in response to short- and long-term environmental changes, but predicting these changes continues to be a challenge. In addition to a chronic lack of data on the cliff face, vegetation presence and growth can bias our erosion measurements and limit our ability to detect geomorphic erosion by obscuring the cliff face. This paper builds on past research segmenting vegetation in three-band red, green, blue (RGB) imagery and presents two approaches to segmenting and filtering vegetation from the bare cliff face in dense point clouds constructed from RGB images and structure-from-motion (SfM) software. Vegetation indices were computed from previously published research and their utility in segmenting vegetation from bare cliff face was compared against machine learning (ML) models for point cloud segmentation. Results demonstrate that, while existing vegetation indices and ML models are both capable of segmenting vegetation and bare cliff face sediments, ML models can be more efficient and robust across different growing seasons. ML model accuracy quickly reached an asymptote with only two layers and RGB images only (i.e., no vegetation indices), suggesting that these more parsimonious models may be more robust to a range of environmental conditions than existing vegetation indices which vary substantially from one growing season to another with changes in vegetation phenology.

Novel View Synthesis with Neural Radiance Fields for Industrial Robot Applications

Abstract. Neural Radiance Fields (NeRFs) have become a rapidly growing research field with the potential to revolutionize typical photogrammetric workflows, such as those used for 3D scene reconstruction. As input, NeRFs require multi-view images with corresponding camera poses as well as the interior orientation. In the typical NeRF workflow, the camera poses and the interior orientation are estimated in advance with Structure from Motion (SfM). But the quality of the resulting novel views, which depends on different parameters such as the number and distribution of available images, the accuracy of the related camera poses and interior orientation, but also the reflection characteristics of the depicted scene, is difficult to predict. In addition, SfM is a time-consuming pre-processing step, and its robustness and quality strongly depend on the image content. Furthermore, the undefined scaling factor of SfM hinders subsequent steps in which metric information is required. In this paper, we evaluate the potential of NeRFs for industrial robot applications. To start with, we propose an alternative to SfM pre-processing: we capture the input images with a calibrated camera that is attached to the end effector of an industrial robot and determine accurate camera poses with metric scale based on the robot kinematics. We then investigate the quality of the novel views by comparing them to ground truth, and by computing an internal quality measure based on ensemble methods. For evaluation purposes, we acquire multiple datasets that pose challenges for reconstruction typical of industrial applications, like reflective objects, poor texture, and fine structures. We show that the robot-based pose determination reaches similar accuracy as SfM in non-demanding cases, while having clear advantages in more challenging scenarios. We also report results of the novel view quality for different NeRF approaches, showing that an additional online pose refinement may be disadvantageous. Finally, we present first results of applying the ensemble method to estimate the quality of the synthetic novel view in the absence of a ground truth.

Multitemporal Structure-from-Motion: A Flexible Tool to Cope with Aerial Blocks in Changing Mountain Environment

Abstract. The application of Structure-from-Motion (SfM) and Multi-View-Stereo matching with aerial images can be successfully used for deriving dense point clouds to analyse changes in the mountain environment, which is characterized by changes due to the action of natural process. The comparison of multiple datasets requires to setup a stable reference system, task that is generally implemented by means of ground control points (GCPs). On the other hand, their positioning may be sometimes difficult in mountains. To cope with this drawback an approach termed as Multitemporal SfM (MSfM) is presented: multiple blocks are oriented together within a unique SfM project, where GCPs are used in only one epoch for establishing the absolute datum. Accurate coregistration between different epochs depends on the automatic extraction of tie points in stable areas. To verify the application of MSfM in real cases, this paper presents three case studies where different types of photogrammetric data are adopted, including images from drones and manned aircrafts. Applications to glacier and mountain river erosion are entailed.

Modeling the Effect of Vegetation Coverage on Unmanned Aerial Vehicles-Based Object Detection: A Study in the Minefield Environment

An important consideration for UAV-based (unmanned aerial vehicle) object detection in the natural environment is vegetation height and foliar cover, which can visually obscure the items a machine learning model is trained to detect. Hence, the accuracy of aerial detection of objects such as surface landmines and UXO (unexploded ordnance) is highly dependent on the height and density of vegetation in a given area. In this study, we develop a model that estimates the detection accuracy (recall) of a YOLOv8 object’s detection implementation as a function of occlusion due to vegetation coverage. To solve this function, we developed an algorithm to extract vegetation height and coverage of the UAV imagery from a digital surface model generated using structure-from-motion (SfM) photogrammetry. We find the relationship between recall and percent occlusion is well modeled by a sigmoid function using the PFM-1 landmine test case. Applying the sigmoid recall-occlusion relationship in conjunction with our vegetation cover algorithm to solve for percent occlusion, we mapped the uncertainty in detection rate due to vegetation in UAV-based SfM orthomosaics in eight different minefield environments. This methodology and model have significant implications for determining the optimal location and time of year for UAV-based object detection tasks and quantifying the uncertainty of deep learning object detection models in the natural environment.

Mapping fault geomorphology with drone-based lidar

The advent of sub-meter resolution topographic surveying has revolutionized active fault mapping. Light detection and ranging (lidar) collected using crewed airborne laser scanning (ALS) can provide ground coverage of entire fault systems but is expensive, while Structure-from-Motion (SfM) photogrammetry from uncrewed aerial vehicles (UAVs) is popular for mapping smaller sites but cannot image beneath vegetation. Here, we present a new UAV laser scanning (ULS) system which overcomes these limitations to survey fault-related topography cost-effectively, at desirable spatial resolutions, and even beneath dense vegetation. In describing our system, data acquisition and processing workflows, we provide a practical guide for other researchers interested in developing their own ULS capabilities. We showcase ULS data collected over faults from a variety of terrain and vegetation types across the Canadian Cordillera and compare them to conventional ALS and SfM data. Due to the lower, slower UAV flights, ULS offers improved ground return density (~260 points/m2 for the capture of a paleoseismic trenching site and ~10–72 points/m2 for larger, multi-kilometer fault surveys) over conventional ALS (~3–9 points/m2) as well as better vegetation penetration than both ALS and SfM. The resulting ~20–50 cm-resolution ULS terrain models reveal fine-scale tectonic landforms that would otherwise be challenging to image.

DeepSFM: Robust Deep Iterative Refinement for Structure From Motion.

Structure from Motion (SfM) is a fundamental computer vision problem which has not been well handled by deep learning. One of the promising solutions is to apply explicit structural constraint, e.g., 3D cost volume, into the neural network. Obtaining accurate camera poses from images alone can be challenging, especially with complicated environmental factors. Existing methods usually assume accurate camera poses from GT or other methods, which is unrealistic in practice and additional sensors are needed. In this work, we design a physical driven architecture, namely DeepSFM, inspired by traditional Bundle Adjustment, which consists of two cost volume based architectures to iteratively refine depth and pose. The explicit constraints on both depth and pose, when combined with the learning components, bring merit from both traditional BA and emerging deep learning technology. To speed up the learning and inference efficiency, we apply the Gated Recurrent Units (GRUs)-based depth and pose update modules with coarse to fine cost volumes on the iterative refinements. In addition, with the extended residual depth prediction module, our model can be adapted to dynamic scenes effectively. Extensive experiments on various datasets show that our model achieves state-of-the-art performance with superior robustness against challenging inputs.

Mapping and quantifying medium-term soil loss rates in mountain olive groves using unmanned aerial vehicle technology

Olive groves are one of the main agroecosystems in the Mediterranean region, but water erosion, aggravated by inappropriate soil management, is compromising the environmental sustainability of these crops. National and international public organisations, including the European Union via its Common Agricultural Policy, have acknowledged the problem and recognise the need to quantify the effects of this process. However, the variability of currently available short-term soil erosion measurements, together with limited understanding of the underlying processes, mean there is considerable uncertainty about the long-term effects of soil erosion. This paper presents an innovative procedure called SERHOLIVE4.0 designed to measure and model long-term soil erosion rates in olive groves, by means of structure-from-motion (SfM) techniques by which image information is obtained from unmanned aerial vehicles (UAVs). For the present study, the procedure was evaluated in mountain olive groves, where the erosion rate was calculated from historical surface reconstructions. Overall, this approach was found to be practical and effective. The method includes the following steps: [1] measure the current relief using UAV technology; [2] reconstruct the historical relief from field measurements; [3] calculate soil truncation (h) and obtain a soil erosion rate map; [4] determine the erosive dynamics of the slope and establish the relation between tree truncation, slope and mounds. The method we describe presents the following advantages:•it quantifies soil losses by reference to existing tree mounds;•it is straightforward to apply;•its application enhances the calibration of erosion models.

Evaluation of 2D affine — hand-crafted detectors for feature-based TLS point cloud registration

Abstract The development of modern surveying methods, particularly, Terrestrial Laser Scanning (TLS), has found wide application in protecting and monitoring engineering and objects and sites of cultural heritage. For this reason, it is crucial that several factors a˛ecting the correctness of point cloud registration are considered, including the correctness of the distribution of control points (both signalised and natural), the quality of the process, and robustness analysis. The aim of this article is to evaluate the quality and correctness of TLS registration based on point clouds converted to raster form (in spherical mapping) and hand-crafted detectors. The expanded Structure-from-Motion (SfM) was used to detect the tie points for TLS registration and reliability assessment. The results demonstrated that affine detectors are useful in detecting a high number of key points (increased for point detectors by 8–12 times and for blob detectors by about 10–24 times), improving the quality and TLS registration completeness. For the registration accuracy of point cloud on signalised check points, the lower values can be noted for maximum RMSE errors for blob affine detectors than detectors and larger values for corner detectors and affine detectors (not more than 4 mm in the extreme cases, typically 2 mm). The commonly-applied target-based registration method yields similar results (di˛erences do not exceed – in extreme cases – 3.5 mm, typically less than 2 mm), proving that using affine detectors in the TLS registration process is and reasonable and can be recommended.

A Robust Mismatch Removal Method for Image Matching Based on the Fusion of the Local Features and the Depth

Feature point matching is a fundamental task in computer vision such as vision simultaneous localization and mapping (VSLAM) and structure from motion (SFM). Due to the similarity or interference of features, mismatches are often unavoidable. Therefore, how to eliminate mismatches is important for robust matching. Smoothness constraint is widely used to remove mismatch, but it cannot effectively deal with the issue in the rapidly changing scene. In this paper, a novel LCS-SSM (Local Cell Statistics and Structural Similarity Measurement) mismatch removal method is proposed. LCS-SSM integrates the motion consistency and structural similarity of a local image block as the statistical likelihood of matched key points. Then, the Random Sampling Consensus (RANSAC) algorithm is employed to preserve the isolated matches that do not satisfy the statistical likelihood. Experimental and comparative results on the public dataset show that the proposed LCS-SSM can effectively and reliably differentiate true and false matches compared with state-of-the-art methods, and can be used for robust matching in scenes with fast motion, blurs, and clustered noise.

Deep learning-based bridge damage cause estimation from multiple images using visual question answering

This paper presents a framework for estimating the cause of damage to bridge members by combining Structure from Motion (SfM) and Visual Question Answering (VQA) techniques. A VQA model was developed that uses bridge images for dataset creation and outputs the damage or member name and its existence based on the images and questions. In the developed model, the correct answer rate for questions requiring the member or damage name were 67.4 and 68.9%, respectively. The correct answer rate for questions requiring a yes/no answer was 99.1%. Based on the developed model, a damage cause estimation method was proposed. In the proposed method, the damage causes are narrowed down by inputting new questions to the VQA model, which are determined based on the surrounding images obtained via SfM and the results of the VQA model. Subsequently, the proposed method was then applied to an actual bridge and shown to be capable of determining damage and estimating its cause. The proposed method could be used to prevent damage causes from being overlooked, and practitioners could determine inspection focus areas, which could contribute to the improvement of maintenance techniques. In the future, it is expected to contribute to infrastructure diagnosis automation.

Efficient Structure from Motion for Large-Size Videos from an Open Outdoor UAV Dataset.

Modern UAVs (unmanned aerial vehicles) equipped with video cameras can provide large-scale high-resolution video data. This poses significant challenges for structure from motion (SfM) and simultaneous localization and mapping (SLAM) algorithms, as most of them are developed for relatively small-scale and low-resolution scenes. In this paper, we present a video-based SfM method specifically designed for high-resolution large-size UAV videos. Despite the wide range of applications for SfM, performing mainstream SfM methods on such videos poses challenges due to their high computational cost. Our method consists of three main steps. Firstly, we employ a visual SLAM (VSLAM) system to efficiently extract keyframes, keypoints, initial camera poses, and sparse structures from downsampled videos. Next, we propose a novel two-step keypoint adjustment method. Instead of matching new points in the original videos, our method effectively and efficiently adjusts the existing keypoints at the original scale. Finally, we refine the poses and structures using a rotation-averaging constrained global bundle adjustment (BA) technique, incorporating the adjusted keypoints. To enrich the resources available for SLAM or SfM studies, we provide a large-size (3840 × 2160) outdoor video dataset with millimeter-level-accuracy ground control points, which supplements the current relatively low-resolution video datasets. Experiments demonstrate that, compared with other SLAM or SfM methods, our method achieves an average efficiency improvement of 100% on our collected dataset and 45% on the EuRoc dataset. Our method also demonstrates superior localization accuracy when compared with state-of-the-art SLAM or SfM methods.

Evaluating geometric measurement accuracy based on 3D model reconstruction of nursery tomato plants by Agisoft photoscan software

Measuring the vegetation structure of a plant with the help of 3D images is a non-destructive method that can be used to monitor the plant’s status during the growing season. Structure from motion (SfM) is a popular optical distance measurement method that uses 2D images to create 3D models. Identifying factors that influence the precision of SfM is important due to its increasing use in phenotyping. The primary objectives of this study were, i) to scrutinize the impact of imaging parameters—including, color checkerboard location, overlap size of images, and background filtering—on the quality indices of 3D models, and ii) to attain precise measurements of leaf area, stem diameter, and height through the combination of images captured from different camera shooting angles. The three nursery tomato plants were photographed at every five-degree rotation of the turntable, five different locations of color checkerboards, and five camera shooting angles. Results indicated the color checkerboard’s presence and position affect the quality indices of the 3D models. Visually and by the lowest mean keypoint size value, the least noisy 3D model was obtained when the wall of the plant pot was covered with a checkerboard. When the base-to-distance (b/y) ratio as the image overlap index increased from 0.1 to 0.2, the generation time of the 3D models decreased by more than two times. Optimal Mean Absolute Percentage Error (MAPE) values of 1.08 %, 1.01 %, and 1.38 % were achieved for the estimation of leaf area, stem diameter, and height, respectively. This was recorded by using images taken at 0°, 30°, and 105° vertical azimuth angles. Constructing a 3D model by integrating images captured from other angles or more azimuth angles represents numerous visual challenges and leads to problems of point mismatch, which affects the accuracy of measurements.