Visualization Of 3D Models Research Articles

Adaptive Deep Learning-Based Point Cloud Geometry Coding

Point clouds are a very rich 3D visual representation model, which has become increasingly appealing for multimedia applications with immersion, interaction and realism requirements. Due to different acquisition and creation conditions as well as target applications, point clouds' characteristics may be very diverse, notably on their density. While geographical information systems or autonomous driving applications may use rather sparse point clouds, cultural heritage or virtual reality applications typically use denser point clouds to more accurately represent objects and people. Naturally, to offer immersion and realism, point clouds need a rather large number of points, thus asking for the development of efficient coding solutions. The use of deep learning models for coding purposes has recently gained relevance, with latest developments in image coding achieving state-of-the-art performance, thus making natural the adoption of this technology also for point cloud coding. This paper presents a novel deep learning-based solution for point cloud geometry coding which is able to efficiently adapt to the content's characteristics. The proposed coding solution divides the point cloud into 3D blocks and selects the most suitable available deep learning coding model to code each block, thus maximizing the compression performance. In comparison to the state-of-the-art MPEG G-PCC Trisoup standard, the proposed coding solution offers average quality gains up to 4.9 and 5.7 dB for PSNR D1 and PSNR D2, respectively.

A 3D Image Reconstruction Model for Long Tunnel Geological Estimation

Long tunnels often collapse during the construction period. To ensure personnel safety, the geological characteristics must be predicted before tunnel face excavation. In this study, the ground-penetrating radar (GPR) technique is introduced to obtain information regarding the tunnel excavation face at a certain interval. The amplitude of the radar echo signal is expressed as a function of the position and travel time. A B-scan strategy is selected for the GPR to obtain tunnel information. A frequency-domain ( w -k) focusing algorithm, namely, a synthetic aperture radar focusing algorithm, is applied to focus scattered radar signals to obtain focused images. A low-pass filter is designed to remove noises from the original signals. The contours of target objects are extracted from the background information using the edge detection technique. Space coordinate values of the objects are converted to polar coordinates using the Hough transform algorithm for 3D modeling. Visual C++ and AutoCAD are combined to develop a 3D CAD model to help managers in controlling the construction process. The system creates 3D visualization model images and evaluates the geological characteristics of the tunnel excavation faces. The Taigu Tunnel located in the Shanxi Province of China is taken as a case study. A procedure for the geological analysis of this tunnel is introduced in detail, and a 3D image model is built. The results show that the 3D model can help predict rock compositions and locate potential hazards. Moreover, it has better accuracy than conventional models and can be applied to similar transportation construction projects.

Three-Dimensional Determination of the Fusion Zone between the Distal Maxilla and the Pterygoid Plate of the Sphenoid Bone and Considerations for Implant Treatment Procedure

During pre-operation planning, an implantologist has to decide about the location of a dental implant based on the available bone, anatomical structures and future prosthetics. The aim of this study was to provide an overview of the configurations of the junction zone of the pterygoid process, maxillary tuberosity and pyramidal process among the population and to determine the usefulness of 3D model visualization in regard to precision of anatomical structure projections for clinical planning. A total of 72 cases were analyzed for seven measurements (lateral, medial, rostral, caudal, area, line-1 longitudinal, line-2 transverse) on both body sides—right (R) and left (L). In 57 cases, age and sex of the patient were given. In 15 cases this information was missing. Among the group of 57 cases with complete data, there were 30 females (F) and 27 males (M). A total of 57 models of upper jaws including the adjacent pterygoid process of the sphenoid bone were taken for investigation. The results of the comparison between the right and left side showed no differences (p > 0.05) in values of the measured parameters. The results of the comparison between males and females showed a statistically significant difference when assessing the line-2 transverse (p < 0.05)—in the male group the average was 8.22 mm, in the female group the average was lower (7.83 mm). No statistically significant differences in values of the measured parameters for females and males were found for the left side nor for the right side. In all examined specimens there was enough bone surface in the fusion zone to allow for the stable placement of one tuberopterygoid implant.

Virtual Reality for urban planning. The port of Palermo: past, present, future.

Virtual reality is proving to be an interesting tool for the visualization of 3d models developed for urban planning. It allows not only interactive visualization, but also the possibility of inserting analogical and informative contents. The case study relates to the development of a 3D model viewable in virtual reality of the Port of Palermo area, of great urban, historical and logistical interest. The system allows the visualization of its current state, historical emergencies and future developments, as well as the comparison between existing services and those planned in the reconfiguration project.

Visual attention as a model for interpretable neuroimage classification in dementia

AbstractBackgroundDeep learning has the potential to aid clinical decision‐making in dementia, by automatically classifying brain images. However, several key limitations currently prohibit clinical adoption: 1) network design must be optimised for 3D neuroimaging; 2) analysis must be computationally feasible; 3) model decisions must be interpretable. Interpretability is particularly crucial, as clinicians need to understand how and why each automated decision is made.MethodWe address these issues using a 3D recurrent visual attention model tailored for neuroimaging: NEURO‐DRAM. The model comprises an agent which, trained by reinforcement learning, learns to navigate through volumetric images, selectively attending to the most informative regions for a given task. We trained and tested NEURO‐DRAM using T1‐weighted MRIs from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset. This entailed n=162 Alzheimer’s Disease (AD) patients and n=160 healthy controls (HCs), split into training (90%) and testing (10%) data. Classification generalisability was evaluated using independent AD patients (n=130) and HCs (n=100) data from the Open Access Series of Imaging Studies (OASIS) dataset. Finally, we assessed the potential to transfer the classification task (i.e., no extra training needed) to discriminate between the baseline MRIs of people with stable or progressive mild cognitive impairment (MCI).ResultNEURO‐DRAM achieved 98.5% balanced accuracy when classifying AD patients from HCs from ADNI and 99.8% in OASIS, significantly out‐performing a baseline convolutional neural network. When classifying stable versus progressive MCI, accuracy was 77.8%. For each test participant, an individualised trajectory was obtained, depicting the brain regions that were used to make the specific classification (Fig. 1). The regions ‘visualised’ by the model’s trajectories included the hippocampus, parahippocampal gyrus and lateral ventricles. Computation time for training NEURO‐DRAM was substantially faster than the baseline network (10 minutes versus 45 minutes).ConclusionUsing a data‐driven approach, near‐perfect classification of AD patients from HCs can be achieved. To reach this high level of performance, our model learns to ‘visually’ attend to the areas of the brain radiologically associated with AD. Importantly, the neuroanatomical trajectory for each individual run through the analysis can be visualised, providing an intuitive way to interpret how NEURO‐DRAM has reached a classification decision.

DIGITAL MODELS AND 3D VISUALIZATION IN ARCHAEOLOGICAL CONTEXTS. THE SURVEY OF THE TOMBS OF THE KINGS IN THE CITY OF VERGHINA IN MACEDONIA

Abstract. The research presents the results of the non-invasive survey campaign conducted at the archaeological site of the city of Verghina in Macedonia. Through the consolidated processes of the disciplines of representation, such as digital surveying, point cloud, flat surface processing, 3D modeling and multi-resolution visualization, it is proposed a path of knowledge of the city and the finds that are the subject of the research. Of great interest is the operational process illustrated both on the basis of the problems that emerged and the solutions adopted in the survey phases. As part of the study, the structures present in the city of Verghina and in the natural area known as the Great Mound, where the Royal Tombs are located, were investigated. The hill of land 13 meters high and 110 meters wide preserves the remains of the ancient city buried in the second century BC to escape the pillage of enemy armies. The research has dealt with the three structures present, namely the Tomb of Philip II, the Tomb of Alexander IV (son of Alexander the Great) and the Tomb of Persephors. The investigations conducted aim to document the current state of the places through digital surveys and parametric modelling, proposing, in addition, the visualization of 3D images through multi-resolution systems derived from the bases of the survey previously carried out.

CONCEPTUAL MODEL OF 3D ASSET MANAGEMENT BASED ON MYSPATA TO SUPPORT SMART CITY APPLICATION IN MALAYSIA

Abstract. Urbanization is the access to modernization and development around the world. Nowadays, with the current technology development, smart cities are seen as a new approach in urban management and development. 3D asset management is one of the components to support the idea of smart city. 3D asset management is important to assist the monitoring and maintenance of asset in smart city by enabling visualization of 3D models, locating and query in real-time based. In Malaysia, the government is looking seriously at the issues of asset management and maintenance. This is happened because asset management in present day is already moving towards the revolution of smart city but still considered as time consuming and open to human errors as the asset managers or authorities still considering on paper-dependent and manual inspection practise. In the past few years, Malaysia has developing an electronic-based asset management, MySPATA that is made prior to the inefficiency on the asset management system. MySPATA has been introduced as electronic based asset management solution for immovable assets that belong to various department and ministries. However, the creation of MySPATA is considered as bland and time-consuming as its application only storing and displaying asset information. Thus, the implementation of 3D asset management is required for a better and effective management. In this paper, we proposed the conceptual model of 3D asset management by incorporating with the new CityGML standard. The proposed 3D asset management is based on MySPATA module. CityGML plays an important role in demonstrating the 3D asset management for modelling, string and exchanging city models in the international standard. So, the 3D asset management is developed based on MySPATA module that integrated with new CityGML concept. Therefore, with this new approach and concept, the managing of assets will lead for better management and maintenance.

EXPLORATORY STUDY OF 3D POINT CLOUD TRIANGULATION FOR SMART CITY MODELLING AND VISUALIZATION

Abstract. The current trends of 3D scanning technologies allow us to acquire accurate 3D data of large-scale environment efficiently. The 3D data of large-scale environments is essential when generating 3D model is for the visualization of smart cities. For the seamless visualization of 3D model, large data size will be used during the 3D data acquisition. However, the processing time for large data size is time consuming and requires suitable hardware specification. In this study, different hardware capability in processing large data of 3D point cloud for mesh generation is investigated. Light Detection and Ranging (LiDAR) Airborne and Mobile Mapping System (MMS) are used as data input and processed using Bentley ContextCapture software. The study is conducted in Malaysia, specifically in Wilayah Persekutuan Kuala Lumpur and Selangor with the size of 49 km2. Several analyses have been performed to analyse the software and hardware specification based on the 3D mesh model generated. From the finding, we have suggested the most suitable hardware specification for 3D mesh model generation.

3D stem model construction with geometry consistency using terrestrial laser scanning data

ABSTRACT The traditional stem model is inconsistent with the real geometry of the stem. Terrestrial laser scanning (TLS) provides a possibility of constructing a realistic stem model. In this study, we present a 3D stem model, which includes the stem axis curve and stem cross-sectional profile curve, with geometrical consistency and stem parameter retrieval methods using TLS data. The 3D stem axis curve is interpolated based on the geometric central points, which are calculated from stem slices formed by stem cross-sections. From the 3D stem axis curve, the stem shape characteristic is clearly depicted in 3D space. Stem parameters, such as the location of a stem cross-section, stem diameter, height, length, curvature and torsion at any position along the stem, are calculated. The stem cross-sectional profile curve is interpolated based on the stem cross-sectional profile points, which are calculated from stem cross-sectional points by angle simplification. Then, the convex-concave characteristic of the stem cross-section is represented by the stem cross-sectional profile curve, and from this, the integral method for basal area calculation is presented. The presented methods were tested on stem points from different tree species with different shape properties. The feasibility and validity of the 3D stem model was demonstrated by 3D stem model visualization. The root mean square error (RMSE) of the presented stem diameter method was 0.129 cm. Compared with the basal area calculated from the cross-sectional profile curve, the mean absolute percentage error (MAPE) and RMSE values of the traditional method were 8.921% and 49.926 cm2, respectively. The stem parameter retrieval experiment demonstrated the accuracy and practicability of the 3D stem model. Compared with the traditional stem model, the 3D stem model can accurately represent the geometric structure of the stem and provide accurate calculations of stem parameters. It will help improve the applications of TLS in forestry inventories.

Permeability prediction of porous geomaterials subjected to freeze-thaw cycles based on 3D reconstruction technology

This study is helpful to improve the understanding of Engineering leakage mechanism in cold regions. Firstly, computed tomography (CT) tests are carried out on the same sample at 20 °C after 0, 5, 10, 15, 20, 25 FT cycles, and the 3D visualization models of sample after 0, 5, 10, 15, 20, 25 FT cycles are obtained by a volume rendering algorithm. Secondly, the 3D visualization models are imported into Fluent, and the passing fluid is liquid water. The permeability coefficient of sample after 0, 5, 10, 15, 20, 25 FT cycles is predicted by computational fluid dynamics (CFD). In addition, a novel pore extraction (PE) algorithm is proposed to significantly improve the calculation efficiency of permeability prediction. Thirdly, the predicted permeability coefficient (20 °C) is compared with the true permeability coefficient (20 °C). The results show that the method proposed in this paper can well predict the permeability coefficient of the sample. Finally, the effect of FT cycles on the seepage field of sample is analyzed, and the mechanism of the permeability evolution of the sample under FT cycles is studied.

Geological-geophysical models of the crust for the White Sea region

The study of the White Sea region and the adjacent area aimed to model the structure of the crystalline portion of the crust at the contact of the northeastern slope of the Fennoscandian Shield and the Russian Plate. Modeling was based on geological, geophysical and DSS profile data, State Geological Map of the Russian Federation (scale 1:1 000 000), Explanatory Notes to Sheets Q-35, 36, 37, and 38 (third generation, 2009), and Tectonic Map of the White Sea and Adjacent Areas (2012). A model was constructed using GIS INTEGRO software (VNIIgeosystems), specifically its procedures for calculation and visualization of 2D and 3D models showing crust density and magnetic fields. The model of the study area shows the structure of the crust and the characteristics of its horizons.

Soft computing for modeling pipeline risk index under uncertainty

The risk of water pipe failure is deemed one of the significant challenges of the 21st century. The risk analysis and modeling are intricate tasks due to the complexity of the buried piping system, which is nonlinear, dynamic, and includes vast arrays of indicators that cannot be measured meticulously in any conventional metrics. Deterioration indicators extracted from field inspections and/or experts' questionnaires have inherently certain degrees of uncertainty and subjective judgments. Soft computing techniques are capable of coping with the imprecision, uncertainty, and fuzziness of data. The objective of this research is to develop a soft computing approach grounded in a fuzzy inference system (FIS) to enable encoding the deterioration indicators into a risk index while dealing with the ambiguity and imprecision. The approach involves five phases that are based on data from Arequipa City in Peru, simulation and FIS, and supported by 3D schematic representations. Data are channeled to the FIS engine after defining the membership functions and 127 fuzzy rules. Subsequent to successive simulation iterations, aggregation, and defuzzification of the outputs, the indices for risk-of-failure are generated along with 3D visualization models. To ensure the coherency of the proposed model, Monte Carlo simulation is performed in conjunction with sensitivity analysis on the data. The validation results of a sample revealed the efficacy of the model with the Coefficient of Variation and the Mean Standard Error of 0.0296 and 0.03, respectively. The five phases are chained together to enable the model to fulfill its functions.

Potato (Solanum tuberosum L.) tuber-root modeling method based on physical properties.

The development of tuber-root models based on the physical properties of the root system of a plant is a prominent but complicated task. In this paper, a method for the construction of a 3D model of a potato tuber-root system is proposed, based on determining the characterization parameters of the potato tuber-root model. Three early maturing potato varieties, widely planted in Northeast China, were selected as the research objects. Their topological and geometric structures were analyzed to determine the model parameters. By actually digging potatoes in the field, field data measurement and statistical analysis of the parameters were performed, and a model parameter database was established. Based on the measured data, the root trajectory points were obtained by simulating the growth of the root tips. Then MATLAB was used to develop a system that would complete the construction of the potato tuber-root 3D visualization model. Finally, the accuracy of the model was verified experimentally. Case studies for the three different types indicated an acceptable performance of the proposed model, with a relative root mean square error of 6.81% and 15.32%, for the minimum and maximum values, respectively. The research results can be used to explore the interaction between the soil-tuber-root aggregates and the digging components, and provide a reference for the construction of root models of other tuber crops.

The Application of Three-Dimensional Visualization in Preoperative Evaluation of Portal Vein Invasion in Hilar Cholangiocarcinoma.

ObjectiveThis study aimed to investigate the use of three-dimensional visualization for preoperative evaluation of portal vein invasion in hilar cholangiocarcinoma (HCCA).MethodsThis recombination study for preoperative computerized tomography images was completed in 42 patients undergoing radical resection of HCCA combined with hepatectomy. Portal vein invasion with postoperative pathology was used as a gold standard to decide if the diagnosis was correct or not. We compared the sensitivity, specificity, positive predictive value, negative predictive value, and total correctness of radiologists and a three-dimensional (3D) visualization model for the assessment of tumor-caused portal vein invasion.ResultsThe findings for the estimation of portal vein invasion by radiologists based on CT images were as follows: sensitivity = 90.9%; specificity = 83.8%; positive predictive value = 66.7%; negative predictive value = 96.3%; and overall accuracy = 85.7%. The findings for estimation by the 3D visualization model were as follows: sensitivity = 90.9%; specificity = 96.8%; positive predictive value = 90.9%; negative predictive value = 96.8%; and overall accuracy = 90.5%.ConclusionThe positive predictive value of 3D visualization technology in the diagnosis of portal vein invasion is notably superior to that of subjective assessment by radiologists. This technique can thus play a significant role in preventing unnecessary resectioning of non-invaded portal veins and hepatectomy.

Application of 3D visualization and 3D printing in individualized precision surgery for Bismuth-Corlette type Ⅲ and Ⅳ hilar cholangiocarcinoma

To explore the application of 3D visualization and 3D printing in individualized precision surgical treatment of Bismuth-Corlette type Ⅲ and Ⅳ hilar cholangiocarcinoma. We retrospectively analyzed the data of 10 patients with hilar cholangiocarcinoma undergoing surgeries under the guidance of 3D visualization and 3D printing in the Department of Hepatobiliary Surgery, Zhujiang Hospital from May 2016 to March 2019. Thin-section CT data of the patients were collected for 3D reconstruction and 3D printing, and the 3D printed models were used for observing the 3D relationship of tumor with the intrahepatic bile duct, hepatic artery, portal vein and hepatic vein system and for performing preoperative simulated surgery and surgical planning. The 3D printed models were subsequently used for real-time intraoperative navigation to guide surgeries in the operating room. 3D visualization models were successfully reconstructed for all the 10 patients and printed into 3D models. The 3D visualization types in Bismuth-Corlette classification included type Ⅲa (4 cases), type Ⅲb (4 cases), and type Ⅳ (2 cases); 4 patients showed portal vein variation, 3 had hepatic artery variation, and 2 had both portal vein and hepatic artery variations. Two patients were found to have trifurcation type of portal vein variation, one had "I-shaped" variation, and one showed the absence of the right anterior branch of the portal vein; 3 patients had hepatic artery variations with the left hepatic artery originating from the left gastric artery (1 case) and the right hepatic artery originating from the superior mesenteric artery (2 cases). Four patients with type Ⅲb underwent left hepatectomy; 4 with type Ⅲa received right hepatectomy; 1 patient with of type Ⅳ received peripheral hepatic resection and another underwent left hepatectomy. The results of preoperative 3D reconstruction, 3D printed model and preoperative planning were consistent with the intraoperative findings. The operative time was 452±75.12 min with a mean intraoperative blood loss of 356±62.35 mL and a mean hospital stay of 15 ± 4.61 days in these cases. One patient had bile leakage and 3 patients had pleural effusion postoperatively, and they were discharged after drainage and medications. No liver failure or death occurred in these cases perioperatively. 3D visualization and 3D printing can facilitate accurate preoperative assessment, surgical planning and surgical procedure optimization for Bismuth-Corlette type Ⅲ and Ⅳ hilar cholangiocarcinoma to improve surgical safety and reduce surgical risks especially in cases of intrahepatic vascular variations.

A MULTI-PURPOSE CULTURAL HERITAGE DATA PLATFORM FOR 4D VISUALIZATION AND INTERACTIVE INFORMATION SERVICES

Abstract. The already arduous task of collecting, processing and managing heterogeneous cultural heritage data is getting more intense in terms of indexing, interaction and dissemination. This paper presents the creation of a 4D web-based platform as a centralized data hub, moving beyond advanced photogrammetric techniques for 3D capture and multi-dimensional documentation. Precise metric data, generated by a combination of image-based, range and surveying techniques, are spatially, logically and temporally correlated with cultural and historical resources, in order to form a critical knowledge base for multiple purposes and user types. Unlike conventional information systems, the presented platform, which adopts a relational database model, has the following front-end functionalities: (i) visualization of high-resolution 3D models based on distance dependent Level of Detail (LoD) techniques; (ii) web Augmented Reality; and (iii) interactive access and retrieval services. Information deduced from the developed services is tailored to different target audiences: scientific community, private sector, public sector and general public. The case study site is the UNESCO world heritage site of Meteora, Greece, and particularly, two inaccessible huge rocks, the rock of St. Modestos, known as Modi, and the Alyssos rock.

Therapeutic Images of CT Image Analysis Based on 3D Visualization Technology in Patients with Hepatobiliary Stones

Objective: Paper for CT imaging in three-dimensional visualization technology for laparoscopy combined with biliary hard lens therapeutic clinical effect of extrahepatic bile duct stones were evaluated. Methods: Abdominal medical image 3D visualization software for our hospital CT imaging diagnosis and treatment of bile duct stones in 45 patients of bile duct stones three-dimensional visualization of clinical analysis and preoperative planning, and brought it into the operating room three-dimensional visualization model, guiding the implementation of the joint 3D laparoscopic biliary lithotripsy targeted hard lens. At the same time, as well as consistency with the actual hepatolithiasis distribution calculation operative 3D model visualization display; The operative time, bleeding, blood transfusion, stone clearance rate, morbidity and mortality perioperative after review stone recurrence rate was observed. Results: reproducing a three-dimensional patient model visualization liver, intrahepatic vascular anatomy of the liver bile duct stones distributed, real intraoperative and preoperative displayed three-dimensional visualization model consistent, pre-operative manner consistent with preoperative planning. Operation time (125.9±21.2) minutes, blood loss (38.8±8.5) ml, no massive hemorrhage and blood transfusion; MRCP examination by stone clearance rate was 100%; 2 biliary injury, bile leakage 1 complication the rate was 6.7%; There were no perioperative deaths. Stone recurrence two cases, the recurrence rate of 4.4%. Conclusion: The three-dimensional visualization techniques may be implemented hepatolithiasis accurate assessment of preoperative, intraoperative guidance of laparoscopic, endoscopic biliary hard gravel, stone liver resection operation, help to improve stone clearance rate, guarantee operation safety.

3D Perception and Augmented Reality Developments in Underwater Robotics for Ocean Sciences

Purpose of ReviewThis paper addresses the benefits and challenges of mixed reality (MR) for the exploration of deep-sea environments with remotely operated vehicles. The approach is twofold: virtual reality (VR) let the scientist explore the environment via a visual 3D model, overcoming limitations of local perception. Augmented reality (AR) concepts are designed in order to improve environment perception and interaction.Recent FindingsThe key to such concepts is the implementation of 3D visual geo-referenced terrain models from the imaging feedback gathered by the vehicle exploring its unknown surroundings. Image processing, underwater vehicle navigation, and user-friendly displays for robotic intervention are addressed in an integrated concept. A broad development programme carried out at the French Institute for Ocean Science, IFREMER, is described and illustrates technical topics and use cases.Summary3D perception derived from camera vision is shown to enable AR concepts that will significantly improve remote exploration and intervention in unknown natural environments. Cumulative geo-referenced 3D model building is in the process of being taken to reliable functioning in real-world underwater applications, accomplishing a milestone change in the capacity to view and understand the obscure and inaccessible deep-sea world.

Window Detection in Facades Using Heatmap Fusion

Window detection is a key component in many graphics and vision applications related to 3D city modeling and scene visualization. We present a novel approach for learning to recognize windows in a colored facade image. Rather than predicting bounding boxes or performing facade segmentation, our system locates keypoints of windows, and learns keypoint relationships to group them together into windows. A further module provides extra recognizable information at the window center. Locations and relationships of keypoints are encoded in different types of heatmaps, which are learned in an end-to-end network. We have also constructed a facade dataset with 3 418 annotated images to facilitate research in this field. It has richly varying facade structures, occlusion, lighting conditions, and angle of view. On our dataset, our method achieves precision of 91.4% and recall of 91.0% under 50% IoU (intersection over union). We also make a quantitative comparison with state-of-the-art methods to verify the utility of our proposed method. Applications based on our window detector are also demonstrated, such as window blending.

Three-dimensional finite element modeling for evaluation of laryngomalacia severity in infants and children.

This study was performed to investigate the feasibility of using a three-dimensional (3D) finite element model for laryngomalacia severity assessment. We analyzed laryngeal computed tomography images of seven children with laryngomalacia using Mimics software. The gray threshold of different tissues was distinguishable, and a 3D visualization model and finite element model were constructed. The laryngeal structure parameters were defined. The peak von Mises stress (PVMS) value was obtained through laryngeal mechanical analysis. The PVMS values of the laryngeal soft tissue and cartilage scaffolds were independently correlated with disease severity. After stress loading the model, the relationship between laryngomalacia severity and the PVMS value was apparent. However, the PVMS value of laryngeal soft tissue was not correlated with laryngomalacia severity. This study established the efficacy of a finite element model to illustrate the morphological features of the laryngeal cavity in infants with laryngomalacia. However, further study is required before widespread application of 3D finite element modeling of laryngomalacia. PVMS values of the laryngeal cartilage scaffold might be useful for assessment of laryngomalacia severity. These findings support the notion that structural abnormalities of the laryngeal cartilage may manifest as quantifiable changes in stress variants of the supraglottic larynx.