Geometric modeling and processing

The optimization of a galvanic process is a crucial task for many manufacturers in the field of electro-deposition industry. This is true for companies operating in the high fashion field, in which expensive materials are used and reducing material waste is crucial. In this paper, the estimation of the wet surface of small metal parts is treated. In fact, considering a single piece, the amount of material required to guarantee a desired plate thickness is directly proportional to its outer surface. Starting with a rapid overview on other methods to come up with this task, the attention in principally oriented to surface estimation by means of optical scanning. A preliminary test session has been carried out and two main issues arose. The first one is related to resolution and accuracy: due to pieces small dimensions and details, high performances are required to achieve valid results. The second and principal issue is related to the high reflectivity of pieces, even before electroplating. With the aim of avoiding the use of matting paint (which is difficult to remove), the attention has been focused on commercial solution dedicated to jewellery and dentistry fields. Three devices (based on white and blue LED structured light) have been tested on high reflective and specular pieces. From the analyses, only the one based on blue LED technology was able to retrieve high reflective surfaces without matting. Minor issues arose in case of specular surface. The device has been considered suitable for the task.

Keywords: 3D optical scanners , 3D reconstruction , Blue LED structured light , Galvanic , Reverse engineering

The orbital walls and floor are common sites of facial bone fracture and may cause severe functional impairment. The complex geometry of the bony orbit makes anatomical reconstruction extremely challenging, with main issues related to the implant's correct shaping, positioning, and orientation inside the orbital cavity. This study proposes an innovative medical device to place patient-specific implants in fractured eye sockets properly. The device must be used with the developed improved version of a tailored implant shaping mould. The design of the orbital implant positioner followed specific clinical and technical requirements and specifications investigated through the Quality Function Deployment method. The device has been conceived to be simple, economical, capable of managing deantigenated bones or titanium meshes for orbital floor and wall, and reusable multiple times. The positioner consists of two handles hinged together and adequately coupled by a spring to allow the grasping and placing of the implant. Positioner and mould have been manufactured in polyamide using the Selective Laser Sintering technique. The system accuracy assessment resulted in promising outcomes. The mould can precisely shape the implant with a lower than 0.1 mm deviation. The implant positioner can place the implant with a rotation angle around the orbital rim of barely 7.1° and 1.2 mm deviation in the mediolateral direction (no deviations in the anteroposterior and superior-inferior directions occur)

Keywords: Computer-aided design , Craniomaxillofacial surgery , Implant design , Medical devices , Rapid prototyping

These authors presented an automatic computer-based method for morphological feature segmentation and recognition for thoracic and lumbar human vertebrae in a previous paper. The method analyses high-density discretized models by segmentation and recognition rules codifying the vertebra morphology information, which does not change between different subjects. The methodology has been demonstrated to be valid and repeatable in segmenting and recognizing morphological features of vertebrae. The proposed one gives repeatable and reproducible results concerning the traditional manual methods. Nonetheless, the method has been tested only on human lumbar and thoracic vertebrae without distinctive pathologies. This paper aims to extend this methodology for much wider use by analyzing single vertebrae affected by common defectiveness in archaeological and medical fields. The results of the experimentations, analyzed by a skilled anthropologist and radiologist, show that the method correctly segments the analyzed morphological features, also for thoracic and lumbar vertebrae with defectiveness: in particular, defects that alter the shape of features or the symmetry of the vertebra, determine the absence of a feature, or heavily change the spatial distribution of the anterior part respect to the posterior one, have been analyzed.

Keywords: 3D medical image analysis , Feature recognition , Thoracic and lumbar vertebrae analysis , Vertebrae analysis computer-based methods

Snap-fit joints represent a simple, economical and straightforward way of joining two different components. The design of the snap-fit joint is usually performed evaluating peak stresses that must be tolerated by the material without incurring into failure or plastic deformations; in addition, the force needed to join and disassemble parts is estimated in relation to ergonomic issues. Finally, the retention force, that is the force required to start disjoining parts, needs to be estimated. The evaluation of peak stresses or insertion/retention/removal forces is commonly performed through finite element method, having identified the respective deformed configuration. A different approach has been here followed considering that it is not trivial to identify the most critical condition in a full joining/disjoining cycle, when complex geometries are being considered. In detail, the snap joint has been modelled as a multibody model including a flexible body, which replicates the part that undergoes major deflections during the process. The model has been validated against experimental force – time curves, recorded for an existing joint, and it has been used to optimize a parametrised snap-fit design. As a result, the joining force has been reduced up to −84%; the disassembly force has been reduced up to −86% and the retention force has been incremented up to +7%. On the whole, a numerical framework to study these joints has been established, keeping the computational time reasonably low (about 40 min for the entire insertion and removal simulation).

Keywords: Geometrical modelling , Multibody model , Plastic components design , Snap-fit joint , Tolerance analysis

The work here presented is part of a wider research project aimed at extracting and using in industrial applications high level semantic information from 3D product models that are described by means of their boundary representation (B-rep). The specific focus of the paper is the recognition among the components of the CAD model of an assembly of those belonging to some categories of standard parts largely employed in mechanical industry. The knowledge of these components is crucial to understand the structure of mechanical products as they have specific meaning and function. Standard parts follow international standard in shape and dimensions, and also typical mounting schemes exist concerning their use in the product assembly. These distinctive features have been exploited as a starting point to develop a multi-step recognition algorithm. It includes a shape-based and a context-based analysis both relying on the geometric and topological analysis of a CAD model. As already anticipated by Voelcker in his visionary ability to anticipate open challenges, the shape of an object alone is not enough to understand its function. Therefore, context assessment becomes crucial to validate the recognition given by the shape-based step. It allows to uniquely recognize components in mechanical CAD models, by confirming correct results, refusing the false positives, as well as choosing the correct one when the assignment is multiple.

Keywords: 3D part recognition , Assembly analysis , CAD model processing , Standard part

The current study aimed to propose a Deep Learning (DL) based framework to retrieve in real-time the position and the rotation of an object in need of maintenance from live video frames only. For testing the positioning performances, we focused on intervention on a generic Fused Deposition Modeling (FDM) 3D printer maintenance. Lastly, to demonstrate a possible Augmented Reality (AR) application that can be built on top of this, we discussed a specific case study using a Prusa i3 MKS FDM printer. This method was developed using a You Only Look Once (YOLOv3) network for object detection to locate the position of the FDM 3D printer and a subsequent Rotation Convolutional Neural Network (RotationCNN), trained on a dataset of artificial images, to predict the rotations' parameters for attaching the 3D model. To train YOLOv3 we used an augmented dataset of 1653 real images, while to train the RotationCNN we utilized a dataset of 99.220 synthetic images, showing the FDM 3D Printer with different orientations, and fine-tuned it using 235 real images tagged manually. The YOLOv3 network obtained an AP (Average Precision) of 100% with Intersection Over Unit parameter of 0.5, while the RotationCNN showed a mean Geodesic Distance of 0.250 (σ = 0.210) and a mean accuracy to detect the correct rotation r of 0.619 (σ = 0.130), considering as acceptable the range [r − 10, r + 10]. We then evaluate the CAD system performances with 10 non-expert users: the average speed improved from 9.61 (σ = 1.53) to 5.30 (σ = 1.30) and the average number of actions to complete the task from 12.60 (σ = 2.15) to 11.00 (σ = 0.89). This work is a further step through the adoption of DL and AR in the assistance domain. In future works, we will overcome the limitations of this approach and develop a complete mobile CAD system that could be extended to any object that presents a 3D counterpart model.

Keywords: Augmented reality , CAD assistance , Deep learning , Neural network

This paper presents a Generative Design Method (GDM) for highly customised Cultural Heritage applications concerning the exhibition and conservation of pottery. As a fundamental requirement, archaeological finds must be preserved in their structural integrity. Additionally, when present, the exposition supports must be aesthetically pleasant meaning that they must be non-invasive in the field of view of the observer. Furthermore, each artefact presents a unique geometry, hence its supporting structure must be designed accordingly. The proposed GDM considers these requirements, adopting a synergy of CAD, CAE, and optimisation tools. It is developed through two phases. The first phase, P1, concerns with the structural integrity of the fragment. In this phase, a Parametric Modelling approach is chosen for its ease of use both in the Finite Element Analysis evaluations of artefacts and in the design and optimisations of feasible supporting structures. The output of the phase P1 is the optimised configuration of the functional elements of the support ('Ci ') which are the interface region between the support itself and the fragment of pottery. They represent the input of the second phase, P2, that aims to generate lightweight concepts for the complete supporting structure considering the optimal 'Ci ' configuration. During this phase, an aesthetics criterion (related to the minimisation of the support's visibility) is also considered to achieve non-invasive supporting structures. Doing so, the GDM provides informed decisions in the early stages of the design activities with a simulation driven approach oriented to manufacturing. In this way, users are able to focus on design requirements since the concept's variants are generated by means of an optimised configuration of standardised components ('Ci') and obstacle geometries.

Keywords: Cultural Heritage , Generative Design , Parametric Modelling , Parametric Optimisation

Digital human modeling and gait analysis are essential for improving hip replacement surgery (HRS). In this study, Convolution Neural Networks (CNN) are used as a machine learning method to extract the most accurate stick-model from videos captured on a simple camera to represent gait and body components. We developed and tested multiple approaches to create an equitable skeleton model from an image. This process consists of two main parts: defining the joint locations using a CNN network in different architectures, and defining the connections into the final skeletons. A CNN has been trained, validated, and tested using the OpenPose software, which combines two different networks that have been tested on three data-sets for learning and evaluation. The results were satisfactory, but MobileNetV1 was evaluated for optimization of OpenPose computations and definitions. Several hyper-parameters were investigated to provide better representations. As a result of utilizing OpenPose methodology in conjunction with heavily optimized network design and post-processing code, and implementing MobileNet, the proposed solution has provided improved accuracy ratios.

Keywords: 2D Joint profile , CNN , Deep Learning , skeleton feature extraction

Blind and visually impaired people are mostly excluded in enjoying visual artwork yet. Even if the effectiveness of tactile supports has been proven in previous studies, these are difficult to realize, since they are commonly handmade. In this paper, a set of computer-aided interactive tools for a semi-automatic reconstruction of tactile bas-relieves is proposed. Starting from the digital picture of a painting, this set make it possible to retrieve a 2.5D reconstruction of a scene in the form of flat-layered bas relief, which means that the scene is reconstructed solely by means of geometric primitives such planes, cylindrical surfaces, conical surfaces and generic (curve) surfaces. Tools have been specifically thought to obtain tactile bas-relieves of architectural scenes. Unlike typical handmade crafting, the proposed tools do not require specific user skills or training. In fact, user is only asked to select points (i.e., to detect a vanishing point) or segments of the picture to obtain a specific surface. Tools have been designed, optimized, and adopted to realize the tactile bas-relief of the painting Piazza San Francesco (unknown artist, Museo Civico di Arte Antica - Pistoia), within the research activities related to ARTE project (Augmented Readability Tactile Exploration), co-founded by Cassa di Risparmio di Pistoia e Pescia.

Keywords: Blind , Shape from single image , Tactile bas-relief

Additive manufacturing (AM) methods have a growing application in different fields such as aeronautical, automotive, biomedical, and there is a huge interest towards the extension of their use. In this paper, lattice structures for AM are analysed with regards to stiffness and printability in order to verify the suitability for applications where the main requirement of efficiency in terms of stiffness has to be balanced with other needs such as weight saving, ease of manufacturing and recycling of the material. At this aim, lattice structures with high porosity unit cells and large cell size made of a recyclable material were considered with a geometrical configuration allowing 3D printing without any supports. The lattice structures considered were based on body-centred cubic (BCC) and face centred cubic (FCC) unit cell combined with cubic cell. Finally, a multi-morphology lattice structure obtained by mixing different unit cells is also proposed. The lattice structures were modelled and structurally analysed by means of finite element method (FEM), manufactured with a Fusion deposition modelling (FDM) printer and evaluated in relation to printability and dimensional accuracy. The results show that the proposed structure with mixed cells is potentially advantageous in terms of weight saving in relation to the mechanical properties.

Keywords: Additive manufacturing , Geometrical configuration , High porosity , Lattice structure , Supportless 3D printing

The demand for orthodontic and aesthetic treatments, aimed at having healthier teeth and more beautiful smiles, is increasingly growing. The devices on which these treatments are based must be rigorously bespoke for each patient. This is amplifying the need to develop digitized workflows, ranging from scanning to Additive Manufacturing (AM). The present work proposes an alternative workflow for designing and manufacturing orthodontic aligners, also known as clear aligners, starting from the intraoral scanning of the patient's dentition. Orthodontic aligners are an alternative to metal brackets to correct dental malocclusions and they are often preferred by the patients because of their lower impact on facial aesthetics and for their higher comfort. The orthodontic treatments based on the aligners utilize a series of aligners, each one with a geometry slightly different from the previous one. The use of the single aligners is aimed to apply a force to the teeth and gradually aligning them until the end of the treatment. The workflow we propose in the present study is based on the following three main stages: intraoral scanning of the patient's dentition, design of the aligners through a semi-automatic algorithm, and the direct additive manufacturing of the aligners through VAT photopolymerization technique. The possibility to directly additive manufacturing the aligners allows us to rethink the current orthodontic treatments. The aligners geometry can be re-designed, with the possibility of locally manipulating the thickness. This approach would allow the regulation of the amount of force applied locally to the tooth, thus optimizing the treatment and its duration. A feasibility study of the proposed workflow is reported in the present paper, with a focus on the semi-automatic design algorithm and on the additive manufacturing process of the aligners.

Keywords: Additive Manufacturing , Bespoke Medical Devices , Dental Appliances , Design Algorithms for Medical Applications , DfAM

The human hand is a versatile and complex body part. It permits difficult movements with various degrees of precision and force. Several causes can lead to upper limb damage, including musculoskeletal disorders and diseases like stroke. The impairment can affect daily living activities. Patients usually undergo rehabilitation therapy with medical personnel for a long time after the traumatic event. In most cases, they use off-the-shelf medical devices. However, the shape of the upper limbs can differ a lot among people. A bespoke rehabilitative device could provide better comfort and usability, but the design process can be challenging. This work aims to present a digital workflow to generate a 3D virtual reconstruction of the patient's upper limb structure, to be used in the device design. Starting from a 3D scan acquisition of the patient's upper limb, the algorithm allows the creation of a polygonal mesh of the arm and the hand by a semi-automatic procedure. The algorithm uses neural networks' capability to automatically detect the upper limb's landmarks to localize the joints' coordinates. The joints' positions can be used to build a virtual skeleton for a 3D model of a human arm. The mesh of the model is subsequently wrapped around the scan of the real arm. The output consists in the 3D rigged model of the patient's upper limb with a manifold mesh that can be deformed using its virtual skeleton. The results have been assessed with patients who had sports injuries or strokes. The 3D deviations between the scan acquisition of the arm and the resulting model have been evaluated.

Keywords: 3D scan , 3D virtual reconstruction , automatic landmarks detection , neural network , upper limb rigged model

In the field of optical 3D scanning for healthcare applications, low-cost depth cameras can be efficiently used to capture geometry at video frame rates. However, the complete reconstruction of anatomical geometries remains challenging since different scans, collected from multiple viewpoints, must be aligned into a common reference frame. This paper proposes a fully automatic procedure to align scans of the upper limb patient's anatomy. A 3D optical scanner, obtained by assembling three depth cameras, is used to collect upper limb acquisitions. A relevant dataset of key points on the hand and the forearm geometry is then determined and used to automatically obtain a rough 3D alignment of the different scans. Hand key points are identified through a neural network, which works on RGB images captured by the depth cameras; forearm key points are recognized by directly processing the point clouds through a specifically designed algorithm that evaluates the skeleton line of the forearm. The approach was tested on forearm acquisitions, and the results were compared to alternative alignment methodologies.

Keywords: 3D optical scanning , automatic point clouds alignment , depth cameras , upper limb anatomy

The treatment of burn scars is a much discussed and sensitive topic because an improper therapy can have a significant impact on the quality of people's lives. To accurately assess both the health of scars and the outcomes of treatment, the medical evaluation should be based on objective measurements of progression over time. To overcome the limitations of subjective assessment is to leverage, 3D scanning technologies can be used to acquire topological information about the lesions and extract a set of relevant statistical parameters describing them. Accordingly, the present work aims at addressing both efficiency and reliability of a preliminary method based on the objective investigation of the surface topography of burn scars by applying it on several patients of the Meyer Children's Hospital burns department. A commercial 3D scanner is used to acquire 3D data relative to the scars of five patients. By applying a method based on the computational analysis of scan data, a significant number of roughness-related parameters are retrieved. This information is used to create a coherent dataset that allows the severity of burn scars to be inferred objectively. The developed method facilitates the evaluation of treatment efficacy by assessing wound healing during follow-up visits.

Keywords: 3D scan , Burn scar , Surface roughness

Additive Manufacturing (AM) is currently making the relevance of lattice structure solutions increasing, allowing the achievement of high performance/mass ratio, where performance stands for energy absorption, stiffness, and/or insulation. This paper undertakes lattice structure for lightweight design of a horse saddletree. Saddletree is the backbone of a horse saddle, and it is composed of different components. In particular, the spring steel reinforcements inside the saddletree make it the heaviest part of the horse saddle, involving also multiple processes of manufacturing and manual assemblies. This paper aims to lightweight an existing saddletree with a Voronoi lattice solution, reducing several manual assemblies. From the methodological point of view, the lightweight design has been based on a multi-scale approach, carried out via nTopology (static FEA on the original bulk design, implicit geometrical lattice generation from FEA result maps and Boolean operation among lattice results and bulk design implicit model). The original bulk design has been digitally acquired and modeled through Reverse Engineering techniques, so that a specific customized solution may be improved. A final weight reduction of 76.5% is achieved, providing an example of how topological optimization techniques coupled with AM (in particular Powder Bed Fusion technology) may reduce assembly efforts

Keywords: FEA , Horse Saddletree , Reverse Engineering , Topological Optimization , Voronoi Lattice Structure

Medical simulation tools have become increasingly popular in recent years for planning complex and non-routine surgeries. This applies also to autologous ear reconstruction, a procedure that is performed when patients have a complete or partial lack of the outer ear. On the basis of the contralateral ear, the physician sculpts a portion of costal cartilage that is then placed subcutaneously in the auricular region. Simulation tools can help the physician to familiarize with the specific geometries and shapes of the ear to be reconstructed. Making simulation tools requires the 3D acquisition of the ear, which can be done through optical scanners or diagnostic images (CT scan). This work focused on the development of a non-invasive, low-cost scanner for the acquisition of the outer ear. To this end, Intel RealSense D405 cameras were chosen, and a dedicated structure for the acquisition of ear anatomy was designed and manufactured, coupled with management software. The design of the scanner leveraged tests made with the Blender software for the optimization of the position and number of devices, while ensuring complete ear acquisition. Tests allowed the validation of the instrument both qualitatively, indicating the satisfaction of the medical staff in using the scanner, and quantitatively, measuring the deviation of the acquisition from a reference model obtained with professional scanners.

Keywords: 3D Ear Acquisition , CAD Modelling , Optical Scanner , Reverse Engineering

The dynamic behavior of a Powered Two-Wheeler (PTW) is much more complicated than that of a car, which is due to the strong coupling between the longitudinal and lateral dynamics produced by the large roll angles. This makes the analysis of the dynamics, and therefore the design and synthesis of the controller, particularly complex and difficult. In relation to assistance in dangerous situations, several recent manuscripts have suggested devices with limitations of cornering velocity by proposing restrictive models. However, these models can lead to repulsion by the users of PTW vehicles, significantly limiting vehicle performance. In the present work, the authors developed an Advanced Rider-cornering Assistance System (ARAS) based on the skills learned by riders running across curvilinear trajectories using Artificial Intelligence (AI) and Neural Network (NN) techniques. New algorithms that allow the value of velocity to be estimated by prediction accuracy of up to 99.06% were developed using the K-Nearest Neighbor (KNN) Machine Learning (ML) technique.

Keywords: advanced rider assistance systems , k-nearest neighbor , machine learning , maximum cornering velocity , powered two-wheeler dynamic behavior

Industrial engineering applications often require manufacturing large components in composite materials to obtain light structures; however, moulds are expensive, especially when manufacturing a limited batch of parts. On the one hand, when traditional approaches are carried out, moulds are milled from large slabs or laminated with composite materials on a model of the part to produce. In this case, the realisation of a mould leads to adding time-consuming operations to the manufacturing process. On the other hand, if a fully additively manufactured approach is chosen, the manufacturing time increases exponentially and does not match the market's requirements. This research proposes a methodology to improve the production efficiency of large moulds using a hybrid technology by combining additive manufacturing and milling tools. A block of soft material such as foam is milled, and then the printing head of an additive manufacturing machine deposits several layers of plastic material or modelling clay using conformal three-dimensional paths. Finally, the mill can polish the surface, thus obtaining a mould of large dimensions quickly, with reduced cost and without needing trained personnel and handcraft polishing. A software tool has been developed to modify the G-code read by an additive manufacturing machine to obtain material deposition over the soft mould. The authors forced conventional machining instructions to match those of an AM machine. Thus, additive deposition of new material uses 3D conformal trajectories typical of CNC machines. Consequently, communication between two very different instruments using the same language is possible. At first, the code was tested on a modified Fused Filament Fabrication machine whose firmware has been adapted to manage a milling tool and a printing head. Then, the software was tested on a large machine suitable for producing moulds for the large parts typical of marine and aerospace engineering. The research demonstrates that AM technologies can integrate conventional machinery to support the composite materials industry when large parts are required.

Keywords: additive manufacturing , aerospace engineering , Fused Filament Fabrication , G-code , hybrid manufacturing , marine engineering , mould

Describing and supplementing geometric shapes (parts) and layouts (assemblies) with relevant information is key for successful product design communication. 3D annotation tools are widely available in commercial systems, but they are generally used in the same manner as 2D annotations in traditional engineering drawings. The gap between technology and practices is particularly evident in plain text annotations. In this paper, we introduce a functional classification of text annotations to provide an information framework for shifting traditional annotation practices towards the Model-Based Definition (MBD) paradigm. In our view, the current classification of dimensions, tolerances, symbols, notes, and text does not stress the inherent properties of two broader categories: symbols and text. Symbol-based annotations use a symbolic language (mostly standardized) such as Geometric Dimensioning and Tolerancing (GD&T) to provide precise information about the implications of geometric imperfections in manufacturing, whereas notes and text are based on non-standardized and unstructured plain text, and can be used to convey design information. We advocate that text annotations can be characterized in four different functional types (objectives, requirements, rationale, and intent), which should be classified as such when annotations are added to a model. The identification and definition of a formalized structure and syntax can enable the management of the annotations as separate entities, thus leveraging their individual features, or as a group to gain a global and collective view of the design problem. The proposed classification was tested with a group of users in a redesign task that involved a series of geometric changes to an annotated assembly model.

Keywords: Annotations , Model-based definition , Text annotations

The following scientific paper aims to analyze in detail the methodology for reverse engineering of a racing motorcycle connecting rod. The objective is to start with a product available on the market as a spare part, reconstruct its CAD model with a high standard of accuracy, then proceed with lightening modifications to arrive at a new, improved design. The innovative aspect of the procedure lies in the fact that in order to ensure accuracy on the order of a tenth of a millimeter during reconstruction, it was decided to use a FARO articulated arm laser to scan the component's outer surface. By taking advantage of appropriate redesign CAD software (Geomagic Design X), a reconstruction can proceed within the high standard of accuracy imposed. In conclusion, the modifications made through material removal allow an improvement in product efficiency, ensuring high performance.

Keywords: CAD modelling , conrod , high-precision modeling , redesign , reverse engineering