Search

Normalized cross-correlation is the reference approach to carry out template matching on images. When it is computed in Fourier space, it can handle efficiently template translations but it cannot do so with template rotations. Including rotations requires sampling the whole space of rotations, repeating the computation of the correlation each time.This article develops an alternative mathematical theory to handle efficiently, at the same time, rotations and translations. Our proposal has a reduced computational complexity because it does not require to repeatedly sample the space of rotations. To do so, we integrate the information relative to all rotated versions of the template into a unique symmetric tensor template -which is computed only once per template-. Afterward, we demonstrate that the correlation between the image to be processed with the independent tensor components of the tensorial template contains enough information to recover template instance positions and rotations. Our proposed method has the potential to speed up conventional template matching computations by a factor of several magnitude orders for the case of 3D images. [ABSTRACT FROM AUTHOR]

In this paper, we propose a machine learning approach for detecting superficial defects in metal surfaces using point cloud data. We compare the performance of two popular deep learning architectures, multilayer perceptron networks (MLPs) and fully convolutional networks (FCNs), with varying feature sets. Our results show that FCNs (F1=0.94) outperformed MLPs (F1=0.52) in terms of precision, recall, and F1-score. We found that transfer learning with pre-trained models can improve performance when the amount of available data is limited. Our study highlights the importance of considering the amount and quality of training data in developing machine learning models for defect detection in industrial settings with 3D images. [ABSTRACT FROM AUTHOR]

Objectives: The aim of this clinical study was to compare the accuracy of intraoral scan system (IOS) with prefabricated aids and stereophotogrammetry (SPG) compared with open tray implant impression (OI) for complete‐arch implant‐supported fixed dental prostheses (CIFDP). Materials and methods: Patients needing CIFDP were enrolled in this study. OI, reference standard, IOS with prefabricated aids, and SPG were performed for each patient. Distance and angle deviations between all pairs of abutment analogs, root mean square (RMS) errors between the aligned test and reference model, and chairside time were measured. The effect of inter‐abutment distance, jaw (maxilla or mandible), number of implants, and arch length on deviations was analyzed. The mixed effect model was applied to analyze deviations and RMS errors. Results: Fifteen consecutive individuals (6 females and 9 males, 47–77 years old) with 22 arches (9 upper and 13 lower jaws) and 115 implants were included. There was no significant difference in distance deviation comparing SPG and IOS with OI (p >.05). IOS showed a significantly greater angle deviation and RMS errors than SPG (median 0.40° vs. 0.31°, 69 μm vs. 45 μm, p <.01). The inter‐abutment distance was negatively correlated with the accuracy of SPG and IOS (p <.05). The chairside time for IOS, SPG, and OI was 10.49 ± 3.50, 14.71 ± 2.86, and 20.20 ± 3.01 min, respectively (p <.01). Conclusions: The accuracy of SPG and IOS with prefabricated aids was comparable. IOS was the most efficient workflow. [ABSTRACT FROM AUTHOR]

Pavement texture and skid resistance are pivotal surface features of roadway to traffic safety, especially under wet weather. Engineering interventions should be scheduled periodically to restore these features as they deteriorate over time under traffic polishing. While many studies have investigated the effects of traffic polishing on pavement texture and skid resistance through laboratory experiments, the absence of real-world traffic and environmental factors in these studies may limit the generalization of their findings. This study addresses this research gap by conducting a comprehensive field study of pavement texture and skid resistance under traffic polishing in the real world. A total of thirty pairs of pavement texture and friction data were systematically collected from three distinct locations with different levels of traffic polishing (middle, right wheel path, and edge) along an asphalt pavement in Oklahoma, USA. Data acquisition utilized a laser imaging device to reconstruct 0.01 mm 3D images to characterize pavement texture and a Dynamic Friction Tester to evaluate pavement friction at different speeds. Twenty 3D areal parameters were calculated on whole images, macrotexture images, and microtexture images to investigate the effects of traffic polishing on pavement texture from different perspectives. Then, texture parameters and testing speeds were combined to develop friction prediction models via linear and nonlinear methodologies. The results indicate that Random Forest models with identified inputs achieved excellent performance for non-contact friction evaluation. Last, the friction decrease rate was discussed to estimate the timing of future maintenance to restore skid resistance. This study provides more insights into how engineers should plan maintenance to restore pavement texture and friction considering real-world traffic polishing. [ABSTRACT FROM AUTHOR]

The recent main topics in ISO/TC159/SC 4/WG 2 and /WG12, which are in charge of display ergonomics in the International Organization for Standardization (ISO), is ergonomics of head‐mounted displays (HMDs) realizing XR or Cross‐Reality, which has attracted attention in recent years. The paper overviews the major issues related to visual comfortability and Image Safety using HMDs, and reports on the key points for standardization to reduce VR sickness and to maintain comfortable viewing of 3D images. [ABSTRACT FROM AUTHOR]

Accurate geometric modeling of blood vessel lumen from 3D images is crucial for vessel quantification as part of the diagnosis, treatment, and monitoring of vascular diseases. Our method, unlike other approaches which assume a circular or elliptical vessel cross-section, employs parametric B-splines combined with image formation system equations to accurately localize the highly curved lumen boundaries. This approach avoids the need for image segmentation, which may reduce the localization accuracy due to spatial discretization. We demonstrate that the model parameters can be reliably identified by a feedforward neural network which, driven by the cross-section images, predicts the parameter values many times faster than a reference least-squares (LS) model fitting algorithm. We present and discuss two example applications, modeling the lower extremities of artery–vein complexes visualized in steady-state contrast-enhanced magnetic resonance images (MRI) and the coronary arteries pictured in computed tomography angiograms (CTA). Beyond applications in medical diagnosis, blood-flow simulation and vessel-phantom design, the method can serve as a tool for automated annotation of image datasets to train machine-learning algorithms. [ABSTRACT FROM AUTHOR]

In this paper, an algorithm for macrofractographic study of materials is proposed on the basis of analysis and sampling of functions and means of measurement of a digital optical microscope. It consists of a consistent implementation of methods developed in this work for studying the fracture surface of materials in a 3D image using digital image processing and using automatic lens recognition and proximity functions. To evaluate the applicability of developed algorithms, dispersion-strengthened composites with an aluminum matrix with different particle sizes and volume fractions fractured under tension were studied. On the basis of results of research, features were established in the mechanisms of fracture and locations of foci and crack nucleation zones during tensile tests depending on the amount content of solid phase in the structure of dispersion-hardened composite materials obtained using internal oxidation technology. [ABSTRACT FROM AUTHOR]

This study proposes a virtual reconstruction method for fuzzy 3D images based on visual communication effects to improve recognition ability. This method includes constructing a grid distribution model for 3D images, recombining spatial information features using visual feature extraction, and detecting edges using contour feature extraction. A three-dimensional image visual communication model was established, and adaptive pixel reconstruction was achieved through fuzzy structure reconstruction method. Reconstruct the texture structure and grayscale histogram of 3D images to achieve high-quality virtual reconstruction. The simulation results verified the effectiveness of the method, and the designed simulation platform verified the accuracy and completeness of the method, demonstrating the improvement of existing systems and the potential for advancing the field of visual communication.

Improving the quality of knowledge in geography classes is possible by effectively applying modern educational methods. It makes geography teaching interactive and interesting, increases students' motivation and makes the learning process more effective. Interactive maps, 3D images, websites and mobile application s help to improve the quality of geography knowledge. These apps provide practical examples and visual organization to help students better understand and remember geography concepts. These methods and experiences are an important tool for improving students' geography knowledge and motivation. [ABSTRACT FROM AUTHOR]

Hand gesture recognition is important in human-computer interaction with wide applications in many fields. Different from common hand gesture recognition based on 2D images acquired from RGB camera, the utilization of 3D images provides additional spatial information about the target and attracts more and more attention in hand gesture recognition. However, most 3D images for hand gesture recognition are based on depth maps, which only take the distance information as a channel of 2D images, without taking full use of the 3D information. Besides, greater data volume of 3D images brings challenges to the arithmetic facility of hand gesture recognition. Here, we proposed a point cloud based method for hand gesture recognition. To fully use the 3D information, plane points for template matching were extracted based on their normal distributions, which leads to the average recognition rate over 97%. Pre-classification was implemented to ensure a high-efficient recognition without additional requirements for the computer. The proposed method may provide approach for accurate and efficient hand gesture recognition based on 3D images. [ABSTRACT FROM AUTHOR]

Training and comparing different images using deep learning techniques play a vital role in image classification. However, these are typically dependent on large scales of datasets and accuracy produced by the models within a certain time. The goal of this paper is to increase accuracy to improve Pneumonia Discovery and Classification by adding batch normalization in a fully connected layer and decreasing processing time by modifying the learning Adam optimizer. The proposed method consists of Convolution neural networks with enhanced fully connected layers and modified learning rate for Adam optimizer to enhance the accuracy and to reduce processing time. Models were trained by passing the input to the different layers of the models. Then the input and concealed (hidden) layer, ReLU is castoff as an initiation function, however, in the production layer softmax is used as a stimulation function which changes the probability distribution of the prediction. For overfitting reduction, dropout is used in individually concealed neurons through 0.5 possibilities. The results were obtained after compiling the model with Adam optimiser for loss function and fitting it with the training and validation datasets. 'Test', 'training', and 'validation' datasets found under 'chest x-ray', consisting of normal class and pneumonia class were used to learn, examine, and legalize the models. Both the proposed models, VGG16 and VGG19, increased the validation accuracy from 87.28% to 98.72% and 88.46 to 92.81% respectively, and the processing time decreased from 30 to 20 ms against a state-of-the-art solution. [ABSTRACT FROM AUTHOR]

ABSTRACT: Objective: This in vitro study aimed to investigate enamel wear against 3 monolithic ceramics using 2 methods of measurement. Methods: Three groups of standard flat monolithic ceramic specimens including zirconia-reinforced lithium silicate glass (Vita Suprinity, VITA Zahnfabrik), yttria-stabilised tetragonal zirconia (Lava Esthetic Zirconia), and lithium disilicate glass (IPS e.max Press, Ivoclar Vivadent) were prepared, with human enamel used as the control group. Each specimen was subjected to the 2-body wear test at 49 N for 250,000 cycles. Enamel antagonists were evaluated with micro-computed tomography (CT) and intra-oral scanner, allowing 3-dimensional images of vertical wear and volumetric loss of enamel antagonists to be calculated. One-way analysis of variance followed by Student–Newman–Keuls post hoc tests were used to examine the differences in vertical wear/volumetric loss amongst the groups. Paired t tests and intra-class correlations were used to compare vertical wear/volumetric loss between the micro-CT and intra-oral scanner groups. Results: No significant difference in vertical wear was found amongst all groups. More volumetric loss was found in all test groups than in the control group (P < .001), but no significant difference was found amongst the test groups. There was a moderate positive correlation (r = 0.535, P = .033) between the vertical wear and volumetric loss. No significant difference between the 2 methods of measurement was found. Conclusions: Monolithic ceramics induce more enamel wear than natural teeth. Both micro-CT and intra-oral scanners can be used for measuring tooth wear with similar performance.

Pavement texture and skid resistance are pivotal surface features of roadway to traffic safety, especially under wet weather. Engineering interventions should be scheduled periodically to restore these features as they deteriorate over time under traffic polishing. While many studies have investigated the effects of traffic polishing on pavement texture and skid resistance through laboratory experiments, the absence of real-world traffic and environmental factors in these studies may limit the generalization of their findings. This study addresses this research gap by conducting a comprehensive field study of pavement texture and skid resistance under traffic polishing in the real world. A total of thirty pairs of pavement texture and friction data were systematically collected from three distinct locations with different levels of traffic polishing (middle, right wheel path, and edge) along an asphalt pavement in Oklahoma, USA. Data acquisition utilized a laser imaging device to reconstruct 0.01 mm 3D images to characterize pavement texture and a Dynamic Friction Tester to evaluate pavement friction at different speeds. Twenty 3D areal parameters were calculated on whole images, macrotexture images, and microtexture images to investigate the effects of traffic polishing on pavement texture from different perspectives. Then, texture parameters and testing speeds were combined to develop friction prediction models via linear and nonlinear methodologies. The results indicate that Random Forest models with identified inputs achieved excellent performance for non-contact friction evaluation. Last, the friction decrease rate was discussed to estimate the timing of future maintenance to restore skid resistance. This study provides more insights into how engineers should plan maintenance to restore pavement texture and friction considering real-world traffic polishing.

Three-dimensional technology using personal data records has been explored for human identification. The present study aimed to explore two methodologies, photography and orofacial scanning, for assessing orofacial records in forensic scenarios, highlighting their impact on human identification. A pilot and quasi-experimental study was performed using Canon 5D-Full Frame equipment (Tokyo, Japan) and an i700 scanner (Medit, Lusobionic, Portugal) (Seoul, Republic of Korea) with Medit Scan for Clinics (MSC) and Smile Design software (V3.3.2). The sample included living patients (n = 10) and individuals in forensic cases (n = 10). The study was divided into two complementary phases: (i) data collection using 2D and 3D technologies and (ii) visual comparison by superimposition procedures, 3D dental images with 3D facial records (3D–3D), and 2D photography with screen printing of 3D facial records (2D-3S). Statistical analyses were performed using descriptive procedures (Likert scale) and the Mann–Whitney U test. The Mann–Whitney U test comparing the data (n = 220 records) from living individuals and those in forensic cases identified statistically significant differences in the performance of the photographic methods for evaluating intraoral mineralisation (p = 0.004), intraoral soft tissues (p = 0.016), intraoral distortion (p = 0.005) and the scan methods for intraoral extra devices (p = 0.003) and extraoral soft tissues (p = 0.005). A visual comparison (n = 40) allowed 3D–3D superimposition. Additionally, 2D-3S superimposition qualitatively identified the middle third of the face as the corporal area within the anatomical features required for successful surgery. In conclusion, the present study presented evidence-based data suggesting that the IO scan method, as an emergent technology, should be explored as a valuable tool in forensic facial identification in real-world scenarios.

Analytic signal is a useful mathematical tool. It separates qualitative and quantitative information of a signal in form of the local phase and local amplitude. Clifford Fourier transform (CFT) plays a vital role in the representation of multidimensional signals. By generalizing the CFT to Clifford linear canonical transform (CLCT), we present a new type of Clifford biquaternionic analytic signal. Due to the advantages of more freedom, the envelop detection problems of 3D images, with the help of this new analytic signal, can get a better visual appearance. Synthesis examples are presented to demonstrate these advantages. [ABSTRACT FROM AUTHOR]

Copyright of China Medical Education Technology is the property of China Medical Education Technology Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Background: Femoral antetorsion in uncemented hip replacement hardly can be modified and the restoration of the anatomic anteversion might be difficult with standard stems. We compared femoral anteversion restoration of a generic straight stem with a proximally fixed anatomic stem that included a dual sagittal curvature and a proximal torsion. It was hypothesised that the restoration of the anteversion was more accurate with the anatomic stem. Patients and methods: In this comparative study data were collected prospectively of 80 consecutive patients with total hip arthroplasty for primary osteoarthritis. In the first 40 patients (Group I) a cementless proximally fixed anatomic stem with 15° antetorsion of the shaft and a dual sagittal curvature was used. Its design was based on a database of 3D CT images of 600 hips. For comparison a cementless generic straight double-tapered stem was implanted in the next 40 patients (Group II). All operations were performed by one experienced surgeon. All patients had a preoperative 3D planning. A low-dose CT scan was performed at 3 months postoperatively to determine the postoperative stem anteversion. Results: The demographics were similar in both groups. In group I the mean postoperative femoral anteversion was similar to the preoperative one (22.1° ± 10.2° vs. 20.4° ± 9°; p = 0.2). In Group II, the mean postoperative femoral anteversion was lower (12.9° ± 10.8°, vs. 18.3° ± 12°; p = 0.02). Conclusions: Uncemented standard femoral stems tend to reduce femoral anteversion. The used anatomic stem restored femoral anteversion better. The clinical impact of this finding has to be proven. [ABSTRACT FROM AUTHOR]

Concerns have recently been expressed about the continuing availability of human bones from India, obtained originally for educational purposes but lacking the requisite informed consent that would be expected today. More generally, a broader claim is being made, namely, that the practice of using any unconsented bones in educational settings is unethical and should cease. These calls, in turn, raise broader issues regarding the availability of anonymous archival collections in anatomy museums. Although this debate centers on undergraduate anatomy teaching, much anthropological research utilizes human remains of past populations for which there can have been no consent. A suggested alternative for undergraduate teaching is the use of 3D images of human bones, rather than the bones themselves. In addressing these issues, the background to the India bone trade is assessed, and the year 1985 is pinpointed as having significant ethical weight. The cultural and ethical interests inherent in studying archival anonymous skeletal material are weighed against indiscriminate reburial. Although any use of unconsented material represents ethical compromise, account should be taken of changing ethical expectations with time. It is concluded that: there is no justification for repatriation or disposal of all bones for which specific informed consent has not been obtained; continued use of anonymous archival human bones in a professional setting is acceptable, even in the absence of informed consent, with the proviso that there are no culturally relevant groups seeking repatriation; the continued existence of bones in long‐standing private collections cannot be justified since it amounts to long‐term storage with no identified goals; the notion that 3D images are an ethically superior alternative to actual human bones is unsustainable, since there is an intimate connection between the bones and the 3D images. [ABSTRACT FROM AUTHOR]

Background: The windlass mechanism (WM) increases the longitudinal arch of the foot via tension of the plantar aponeurosis during dorsiflexion of the metatarsophalangeal (MTP) joint. The purpose of this study was to perform a 3-dimensional evaluation of the displacement of each joint and the height of the navicular during dorsiflexion of the first MTP joint by using weightbearing computed tomography (CT). Methods: Participants were 6 men and 8 women with 23 healthy feet. CT of the foot with a load equivalent to the participant's body weight was performed. The first MTP joint was in the neutral position and dorsiflexed 30 degrees. Between the conditions, we measured the (1) rotation of each bone, (2) rotation of the distal bone with respect to the proximal bone at each joint, and (3) height of the navicular. Results: With respect to the tibia, the calcaneus was at 0.8 ± 0.7 degrees dorsiflexion and 1.4 ± 0.9 degrees inversion, while the talus was at 2.0 ± 1.2 degrees dorsiflexion and 0.1 ± 0.8 degrees eversion. The navicular was at 1.3 ± 1.2 degrees dorsiflexion and 3.2 ± 2.1 degrees inversion, whereas the medial cuneiform was at 0.3 ± 0.6 degrees plantarflexion and 1.3 ± 1.1 degrees inversion. At the talonavicular joint, the navicular was at 0.7 ± 1.3 degrees plantarflexion, whereas at the cuneonavicular joint, the medial cuneiform bone was at 1.4 ± 1.4 degrees plantarflexion. The height of the navicular increased by 1.1 ± 0.6 mm. Conclusion: We 3-dimensionally confirmed the dynamics of WM and found that the calcaneus, navicular, and medial cuneiform moved in all 3 planes. The results suggest that the cuneonavicular joint has the greatest movement among the joints. We believe that these findings will help to elucidate the pathogenesis of WM-related diseases and lead to advances in treatments for pathologies involving the longitudinal arch. Level of Evidence: Level IV, case series. [ABSTRACT FROM AUTHOR]

An operator feels a burden when he/she controls a rescue robot remotely because he/she has to keep watching camera images to find the target object. We think that this burden can be reduced by the combination of Head Mounted Display (HMD) and object recognition by deep learning. In the first half part of this study, we examine the effect that how presentation method by You Only Look Once (YOLO), a deep learning algorithm, and its recognition results to an operator wearing HMD. In the experiment, three methods of presentation were set: no display of object recognition, display only one object recognition result, and display 80 kinds of object recognition results. Under each presentation method, we measured the time it took for the operator to operate the robot and complete the given task. Additionally, we ask a questionnaire for each experiment. The results of the questionnaire showed that the method to present only one object recognition result was useful. In the second half part of this study, we develop a system to present 3D images with YOLO added, to further ease the burden of object search. Furthermore, we numerically prove that this system represents depth. In the experiment, two methods of displaying were set up: 2D images with Bounding Box (BB) by YOLO and 3D images with BB by YOLO. For each method of presentation, the operator operated the robot and recorded the number of objects found within a time limit. Additionally, we asked a questionnaire at the end of the search in each condition and at the end of all the experiments. The results of the questionnaire suggested points that need to be improved. Furthermore, we consider the flicker of the image found in the experiment. [ABSTRACT FROM AUTHOR]

Vascular brain damage can manifest as various ophthalmic pathologies. It is important to understand cerebral circulation to understand the aetiopathogenesis of the disease. It can manifest as various ocular pathologies with visual field defects, cranial nerve palsies, orbital diseases, etc. In this review, with the help of three-dimensional (3D) images, various ophthalmology-related cerebrovascular diseases, their diagnostic methods and their treatment modalities are explained in detail for neophytes.

Aim: Cleft lip and palate (CLP) patients have a greater predisposition to tooth malformation, which could affect pulp volume. The aim of this study is to evaluate the dental pulp volume of central incisors in 3D images between individuals with and without CLP. Materials and Methods: This retrospective case-control study is single-centered and was recruited between January 2016 and October 2022. Ninety-four patients who were followed in the Institute of Orthodontics were evaluated and divided into two groups: a control group of patients without CLP and a test group of patients with CLP. The 3D data were imported by 3D image semi-automatic segmenting software named ITK-SNAP to calculate tooth pulp volume. Results: The dental pulp volume for both groups, control and CLP, did not show statistically significant differences. In the cleft group, when comparing the pulp volume between the cleft side and the non-cleft side, the cleft side showed a smaller volume. Regarding age and sex, no statistically significant differences were observed. Conclusions: Although there are no differences in mean pulp volume between patients with and without CLP, there is a pulp volume reduction in the teeth on the cleft side when compared to the unaffected contralateral side. [ABSTRACT FROM AUTHOR]

Batu Caves is an isolated limestone massif located in the outskirts of north Kuala Lumpur. Dark Cave supports significant species endemism with its diverse habitats. Nine reported caverns have been mapped using conventional speleological methods. The selection of Dark Cave as a Conservation Site was able to limit the number of visitors, preventing intrusive surveys and making non-intrusive methods such as Light Detection and Ranging (LIDAR) scanning useful. In this study, the interior of the cave was mapped using LIDAR to produce a 3D model of the caverns. The results showed that the LIDAR model produced a more accurate representation of Dark Cave than the conventional 2D map. Features such as cavern's dimension, bat roosting locations, and man-made structures revealed in this 3D model are novel from previous conventional maps. Integration of LIDAR-acousticvisualisation survey methods further confirms four bat species (Taphozous melanopogon, Hipposideros diadema, Eonycteris spelaea, Rousettus leschenaultii) roosting in the Dark Cave. T. melanopogon roosts at the brightest part of Dark Cave, while colonies of H. diadema were only found in the pitch-dark areas of Dark Cave. These outcomes produced a more accurate map as reference to scientists and management authority can use it in important conservation decision-making in the future. [ABSTRACT FROM AUTHOR]

Accurate geometric modeling of blood vessel lumen from 3D images is crucial for vessel quantification as part of the diagnosis, treatment, and monitoring of vascular diseases. Our method, unlike other approaches which assume a circular or elliptical vessel cross-section, employs parametric B-splines combined with image formation system equations to accurately localize the highly curved lumen boundaries. This approach avoids the need for image segmentation, which may reduce the localization accuracy due to spatial discretization. We demonstrate that the model parameters can be reliably identified by a feedforward neural network which, driven by the cross-section images, predicts the parameter values many times faster than a reference least-squares (LS) model fitting algorithm. We present and discuss two example applications, modeling the lower extremities of artery–vein complexes visualized in steady-state contrast-enhanced magnetic resonance images (MRI) and the coronary arteries pictured in computed tomography angiograms (CTA). Beyond applications in medical diagnosis, blood-flow simulation and vessel-phantom design, the method can serve as a tool for automated annotation of image datasets to train machine-learning algorithms.

Holography is an optical (interferometric-diffractive) technique that allows the recording and reconstruction of 3D images of objects, as a hologram carries the intensity and phase information of a 'holographed' object. For its study and realization, conventional, theoretical and experimental optical tools are applied. On the other hand, the evolution of computers in data processing and storage, has enabled the viability of computational holograms, the so-called computer generated holograms (CGH). Currently, the experimental implementation of CGHs is done through new opto-electronic devices such as spatial light modulators (SLM), high resolution CCDs cameras, in addition to new photosensitive materials; thus allowing the recording and optical reconstruction of wavefronts of three-dimensional objects and special optical beams. Thus, the purpose of these works is to present and make available to the academic, scientific and other areas of knowledge our studies and algorithms (routines) of computational holographic generation in an accessible and didactic way. Specifically, in this first article, the purpose is to study computer generated holography (CGH) from simplified algorithms (software) of virtual 3D images; as well as, to present simulation results of construction (recording) and numerical reconstruction (reading) of CGHs from simple 2D and 3D images using MATLAB. These present interesting possibilities for experimental realization through their optical reconstruction (reproduction) using SLMs devices to obtain 3D images. [ABSTRACT FROM AUTHOR]

Introduction: The use of automation and sensor-based systems in livestock production allows monitoring of individual cows in real-time and provides the possibility of early warning systems to take necessary management actions against possible anomalies. Among the different RT monitoring parameters, body weight (BW) plays an important role in tracking the productivity and health status.Methods: In this study, various supervised learning techniques representing different families of methods in the machine learning space were implemented and compared for performance in the prediction of body weight from 3D image data in dairy cows. A total of 83,011 records of contour data from 3D images and body weight measurements taken from a total of 914 Danish Holstein and Jersey cows from 3 different herds were used for the predictions. Various metrics including Pearson’s correlation coefficient (r), the root mean squared error (RMSE), and the mean absolute percentage error (MAPE) were used for robust evaluation of the various supervised techniques and to facilitate comparison with other studies. Prediction was undertaken separately within each breed and subsequently in a combined multi-breed dataset.Results and discussion: Despite differences in predictive performance across the different supervised learning techniques and datasets (breeds), our results indicate reasonable prediction accuracies with mean correlation coefficient (r) as high as 0.94 and MAPE and RMSE as low as 4.0 % and 33.0 (kg), respectively. In comparison to the within-breed analyses (Jersey, Holstein), prediction using the combined multi-breed data set resulted in higher predictive performance in terms of high correlation coefficient and low MAPE. Additional tests showed that the improvement in predictive performance is mainly due to increase in data size from combining data rather than the multi-breed nature of the combined data. Of the different supervised learning techniques implemented, the tree-based group of supervised learning techniques (Catboost, AdaBoost, random forest) resulted in the highest prediction performance in all the metrics used to evaluate technique performance. Reported prediction errors in our study (RMSE and MAPE) are one of the lowest in the literature for prediction of BW using image data in dairy cattle, highlighting the promising predictive value of contour data from 3D images for BW in dairy cows under commercial farm conditions.

Deep learning (DL) based semantic segmentation methods have achieved excellent performance in biomedical image segmentation, producing high quality probability maps to allow extraction of rich instance information to facilitate good instance segmentation. While numerous efforts were put into developing new DL semantic segmentation models, less attention was paid to a key issue of how to effectively explore their probability maps to attain the best possible instance segmentation. We observe that probability maps by DL semantic segmentation models can be used to generate many possible instance candidates, and accurate instance segmentation can be achieved by selecting from them a set of “optimized” candidates as output instances. Further, the generated instance candidates form a well-behaved hierarchical structure (a forest), which allows selecting instances in an optimized manner. Hence, we propose a novel framework, called hierarchical earth mover’s distance (H-EMD), for instance segmentation in biomedical 2D+time videos and 3D images, which judiciously incorporates consistent instance selection with semantic-segmentation-generated probability maps. H-EMD contains two main stages: (1) instance candidate generation: capturing instance-structured information in probability maps by generating many instance candidates in a forest structure; (2) instance candidate selection: selecting instances from the candidate set for final instance segmentation. We formulate a key instance selection problem on the instance candidate forest as an optimization problem based on the earth mover’s distance (EMD), and solve it by integer linear programming. Extensive experiments on eight biomedical video or 3D datasets demonstrate that H-EMD consistently boosts DL semantic segmentation models and is highly competitive with state-of-the-art methods. [ABSTRACT FROM AUTHOR]

This paper presents a high-performance 2D-to-3D conversion algorithm technique and its VLSI architecture design for natural scenes. In this study, the depth map was generated based on the horizontal dividing line concept. Also, the dividing line was estimated by obvious color difference on the local region, and the current dividing line was a demarcation from the maximum value to the minimum one for the depth map. A Depth Based Image Rendering technology was employed to generate stereoscopic images according to the image channel of the depth map. Based on the proposed algorithm, a real-time VLSI architecture is presented. Also, for the purpose of cost efficiency, a module-based hardware consisting of a timing schedule control should be designed. The multiplications and divisions of the algorithm can be minimized by circuit design, and so this helps reduce the complexity of this system. Only one frame memory is required to generate real-time 3D images using the depth map. The depth map can be immediately calculated by referring to the location of dividing lines, which can reduce the storage size and I/O bandwidth. The circuit is simulated and verified by one FPGA chip. The critical path is at the multiplex and one flip-flop, and the maximum clock rate can achieve 205 MHz. To be VGA compatible, the stereoscopic RGB pixels can be outputted in parallel per clock to the interface. The overall maximum data rates of the proposed 3D conversion chip can achieve 610 M bytes per second. This can meet the real-time HD requirement. [ABSTRACT FROM AUTHOR]

Background: Patients with cleft lip and palate anomalies suffer from associated negative psychosocial effects despite undergoing reconstructive surgeries. Discrimination from others may be a contributing factor. We aim to understand whether people react differently to cleft and non-cleft subjects in questions regarding aesthetic appraisal and interpersonal impressions, as well as whether diverging opinions exist between people of different backgrounds., Materials and Methods: Using the 3dMD system, three-dimensional craniofacial images of cleft (n = 20) and non-cleft (n = 5) pediatric individuals were obtained and used to create a Likert scale-based questionnaire. Participants from three different backgrounds, including adolescent peers, adult laypersons, and plastic surgeons, were recruited to complete the assessment. The resulting data were transferred to a bar graph and analyzed with the item response theory-graded response model to estimate the latent traits and capability of each subject., Results: Non-cleft images obtained more positive responses than cleft images in questions of both aesthetic appraisal and interpersonal impressions. However, some of the cleft images were rated similarly, or even superior, to non-cleft images in questions regarding interpersonal impressions. Surgeons showed the greatest latent capability in differentiating between cleft and non-cleft images, followed by adult laypersons, and then adolescent peers., Conclusions: Despite cleft patients experiencing drastic improvements in appearance post-repair, all subject groups responded differently to questions concerning aesthetic and social impressions of pediatric cleft and non-cleft images. Plastic surgeons were more critical than others. These findings may offer further insight to clinicians, educators, and caregivers into the social challenges faced by cleft individuals., (Copyright © 2024 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.)

Aim: Cleft lip and palate (CLP) patients have a greater predisposition to tooth malformation, which could affect pulp volume. The aim of this study is to evaluate the dental pulp volume of central incisors in 3D images between individuals with and without CLP. Materials and Methods: This retrospective case-control study is single-centered and was recruited between January 2016 and October 2022. Ninety-four patients who were followed in the Institute of Orthodontics were evaluated and divided into two groups: a control group of patients without CLP and a test group of patients with CLP. The 3D data were imported by 3D image semi-automatic segmenting software named ITK-SNAP to calculate tooth pulp volume. Results: The dental pulp volume for both groups, control and CLP, did not show statistically significant differences. In the cleft group, when comparing the pulp volume between the cleft side and the non-cleft side, the cleft side showed a smaller volume. Regarding age and sex, no statistically significant differences were observed. Conclusions: Although there are no differences in mean pulp volume between patients with and without CLP, there is a pulp volume reduction in the teeth on the cleft side when compared to the unaffected contralateral side.

Most commercially successful face recognition systems combine information from multiple sensors (2D and 3D, visible light and infrared, etc.) to achieve reliable recognition in various environments. When only a single sensor is available, the robustness as well as efficacy of the recognition process suffer. In this paper, we focus on face recognition using images captured by a single 3D sensor and propose a method based on the use of region covariance matrixes and Gaussian mixture models (GMMs). All steps of the proposed framework are automated, and no metadata, such as pre-annotated eye, nose, or mouth positions is required, while only a very simple clustering-based face detection is performed. The framework computes a set of region covariance descriptors from local regions of different face image representations and then uses the unscented transform to derive low-dimensional feature vectors, which are finally modeled by GMMs. In the last step, a support vector machine classification scheme is used to make a decision about the identity of the input 3D facial image. The proposed framework has several desirable characteristics, such as an inherent mechanism for data fusion/integration (through the region covariance matrixes), the ability to explore facial images at different levels of locality, and the ability to integrate a domain-specific prior knowledge into the modeling procedure. Several normalization techniques are incorporated into the proposed framework to further improve performance. Extensive experiments are performed on three prominent databases (FRGC v2, CASIA, and UMB-DB) yielding competitive results. [ABSTRACT FROM AUTHOR]