Image Processing & Applications

Cellular materials have a bulk matrix with a larger number of voids named also cells. Metallic foams made by powder technology represent stochastic closed cells. The related inhomogeneity leads to a scattering of results both in terms of stress-strain curves and maximum strength. Scattering is attributed to relative density variations and local cell discontinuities and it is confirmed also in case of dynamic loading. Finite element simulations through geometrical models that are able to capture the void morphology (named “mesoscale models”), confirm these results and some efforts have been already done to quantify the relationship between shape irregularities and mechanical behavior. The aim of this paper is to present the dynamic characterization of an AA7075 closed cell material and to calibrate its mesoscale finite element model according to the related cell shape distribution. Specimens have been derived from a small ingot (45x45x100 mm) divided along sections so that morphological analysis and experimental tests have been carried out. Specimens extracted from a half of the ingot have been used for dynamic compression tests by means of a split Hopkinson bar, meanwhile specimens extracted from the other half of the ingot have been dissected for porosity distribution analyses carried out by means of image analysis. Stress-strain curves obtained from the mechanical tests have been discussed in terms of strain rate and statistical descriptors of the porosity. Successively a 3D-model of the specimen has been generated starting from the Voronoi algorithm, assigning as input the above-mentioned statistical distribution of the porosity. Due to the peculiarity of the cell morphology (e.g. single larger cells), stress-strain localization has been demonstrated as one of the reasons of the scattering found during the experiments. A material model, to reproduce the investigated foam mechanical behavior, has been calibrated. Despite the difference among experiments the material model is able to reproduce all of them. Difference between the model coefficients quantifies roughly the difference due to the local geometry of the cells.

Keywords: Aluminum foams , Image analysis , Impact test , Porosity distribution , Strain rate effect , Voronoi model

The introduction of the vacuum bell (VB) for the conservative treatment of Pectus Excavatum (PE) has led to a new non-invasive alternative to thoracic surgery. The VB works by elevating the chest as long as a negative differential pressure is internally assured. In recent years studies have been conducted to validate this type of treatment and to outline its correct use; results show a short-term PE improvement when the device is worn for a minimum of 30 min (twice a day) up to a maximum of several hours a day for 12–15 months. Although the worldwide diffusion of VB devices increases year after year, its ability to lift the chest during treatment with respect to the applied pressure has begun to be evaluated only recently. In this paper, a new instrument for measuring chest elevation during treatment is presented and validated. The proposed system consists of two measurement devices: a commercial instrument for the detection of the negative pressure inside the VB, and a specifically developed optical system for the detection of chest movement. The effectiveness of the proposed system, tested on five paediatric patients, paves the way to the objective definition of an optimised patient specific VB scheme of use.

Keywords: Image processing , Optical measurement , Pectus Excavatum , Vacuum bell

Additive Manufacturing is becoming a suitable production process for many industries: it is based on the idea of adding material layer by layer, in opposite to traditional manufacturing processes. This technology shows advantages as design flexibility, internal logistics minimization and product customization that make it perfect to produce customized parts and all the applications where low production rates occur. The production of spare parts for classic or luxury cars is a field where Additive Manufacturing can be adopted because of low demand and relevant costs to manage stocks keeping several different parts in the after-sales inventory. The photogrammetry technique has been investigated to obtain the 3D model of the component to be replaced and send it to decentralized production centers equipped with 3D printers. This approach can enhance by far the supply chain management for automotive spare parts.

Keywords: Additive Manufacturing , Automotive , Maintenance , Photogrammetry , Supply chain management

Nowadays, the development of Internet of Things (IoT) technologies have been enhancing the factory digitalization with several advantages in terms of production efficiency, product quality, and cost reduction. This opportunity encourages the implementation of digital twins related to physical systems for controlling the production workflow in real time. Firstly, the paper studies the enabling technologies for supporting the defect analysis in the context of Industry 4.0 for mechanical workpieces. Secondly, the approach aims to study the integration between the CAD geometry and the quality check process for the inspection planning. A Knowledge-Based tool has been proposed to support the configurations of the quality control chain for each CAD geometry. The test case is focused on the fragmented production of customized gearbox parts.

Keywords: Gearbox , Industry 4.0 , Knowledge Base , Machine Vision , Quality Control

Background: Lumbrical muscles originate in the palm from the 4 tendons of the flexor digitorum profundus and course distally along the radial side of the corresponding metacarpophalangeal joints, in front of the deep transverse metacarpal ligament. The first and second lumbrical muscles are typically innervated by the median nerve, and third and fourth by the ulnar nerve. A plethora of lumbrical muscle variants has been described, ranging from muscles' absence to reduction in their number or presence of accessory slips. The current cadaveric study highlights typical and variable neural supply of lumbrical muscles. Materials: Eight (3 right and 5 left) fresh frozen cadaveric hands of 3 males and 5 females of unknown age were dissected. From the palmar wrist crease, the median and ulnar nerve followed distally to their terminal branches. The ulnar nerve deep branch was dissected and lumbrical muscle innervation patterns were noted. Results: The frequency of typical innervations of lumbrical muscles is confirmed. The second lumbrical nerve had a double composition from both the median and ulnar nerves, in 12.5% of the hands. The thickest branch (1.38 mm) originated from the ulnar nerve and supplied the third lumbrical muscle, and the thinnest one (0.67 mm) from the ulnar nerve and supplied the fourth lumbrical muscle. In 54.5%, lumbrical nerve bifurcation was identified. Conclusion: The complex innervation pattern and the peculiar anatomy of branching to different thirds of the muscle bellies are pointed out. These findings are important in dealing with complex and deep injuries in the palmar region, including transmetacarpal amputations.

Keywords: anomaly , clinical implication , lumbrical muscle , lumbrical nerve , variation

In the field of Engineering, research has conveniently exploited the fluids for energy production. The possibility to use marine renewable energy is still under development, in particular, among the wave energy converter devices the U-OWC systems are the most promising. The main objective of this work is to validate a numerical model with an experimental campaign that aims to simulate the flow field in front of the breakwater and inside the U-OWC. The tests were carried out to understand the hydrodynamic behaviour of the device in regular wave conditions, inside a flume with rectangular section, equipped by a piston-type wave-maker and a U-OWC device, reproducing the REWEC caisson installed in the Natural Ocean Engineering Laboratory (NOEL) of Reggio Calabria, with a 1:13.5 scale. Measurements of the water free surface were used exclusively to validate the 2D numerical model developed through the Ansys Fluent Computational Fluid-Dynamics (CFD) Software. The numerical model solves the fluid flow field using the RANS equations, in which the air-water interaction governed by this set of partial difference equations is solved with the Finite Volume Method (FVM). In conclusion, results related to the energy efficiency of the caisson were extrapolated from the validated numerical model.

Keywords: Computational fluid dynamics , Image analysis , U-OWC , Wave energy

The present work shows how 3D models extracted from a computerised tomography (CT) scan can be processed to be 3D printed into 1:1 orthopedic scale models, which find unquestionable utility in pre-operative surgical planning. Relying on the CAT-CAD methodology, which produces a 3D surface called “mesh” from diagnostic images of body parts, the CAD-AM process elaborates a volumetric bone model which a cost-efficient FDM printer can work with. The suitable materials for these applications are PLA polymers, due to their thermo-mechanical properties, affordability and ecological sustainability; these anatomic 3D printed models allows surgeons to accurately see bones injuries and trauma, resulting in a minimisation of risk and a much more flowing doctor-patient communication. Furthermore these 3D printed objects can be manufactured with specific density in order to simulate bone tissues, resulting in a useful tool through which experienced surgeons can pass on their knowledge to medical students at a very reasonable cost, overcoming the glaring limitations of two-dimensional images provided by CT scans. Here represented is a 3D printed 1:1 scale model of a femur donated to the Bone Bank of IOR-Rizzoli Orthopaedic Institute in Bologna.

Keywords: CT scan , FDM printer , Orthopaedic Reproduction Model , PLA polymer

Pectus Excavatum (PE) is a congenital anomaly of the ribcage, at the level of the sterno-costal plane, which consists of an inward angle of the sternum, in the direction of the spine. PE is the most common of all thoracic malformations, with an incidence of 1 in 300-400 people. To monitor the progress of the pathology, severity indices, or thoracic indices, have been used over the years. Among these indices, recent studies focus on the calculation of optical measures, calculated on the optical scan of the patient's chest, which can be very accurate without exposing the patient to invasive treatments such as CT scans. In this work, data from a sample of PE patients and corresponding doctors' severity assessments have been collected and used to create a decision tool to automatically assign a severity value to the patient. The idea is to provide the physician with an objective and easy to use measuring instrument that can be exploited in an outpatient clinic context. Among several classification tools, a Probabilistic Neural Network was chosen for this task for its simple structure and learning mode.

Pectus Excavatum (PE) is a congenital anomaly of the thoracic cage, at the level of the sternal-costal plane, which consists of an inward angle of the sternum, in the direction of the spine. Its incidence, equal to 1 in 300–400 people, makes it the most frequent among thoracic malformations. For a proper care of patients suffering from PE, it is essential to monitor the progression of the disease and estimate its severity, in order to follow the course of the pathology over time and define the correct treatment. To this end, over the years, severity indexes, or chest indexes, have been introduced. Among them, in recent years, the so-called optical indices, calculated on optical scans of the patient's chest, are gaining ground. In fact, the optical scan is faster to apply and prevents the patient from being exposed to radiation. In this work the peculiar morphological features of PE have been examined by extracting various geometric parameters, in order to develop an outpatient support tool for the evaluation of the severity of the pathology and the monitoring of its progression over time. The extracted values have been compared against a ground truth obtained through five independent surveys collected from paediatric specialists. A Linear Discriminant Analysis was performed to determine the accuracy of the classification using the proposed geometric parameters, obtaining positive results.

Keywords: Clinical support , Depth camera , Pectus Excavatum , Severity index

The most common clinical treatment for ear deformities or non-congenital abnormalities is the reconstruction of the missing geometry using autologous costal cartilage. The surgical procedure consists in cutting, sculpting and suturing harvested costal cartilage from the patient to recreate an ear shape which is symmetric to the contralateral ear. During chirurgical operation, surgeons needs an accurate 3D template as reference to reproduce the ear. For this purpose, reverse engineering and additive manufacturing techniques can be employed. Specifically, this works aims to develop a reliable, low-cost and user-friendly system, to acquire the healthy ear geometry in clinical environment avoiding head patient's exposition to radiation (MRI, CT scan). An ideal acquisition setup and device have been selected to achieve accurate results. To this end, a casted model of an ear was created as reference, and the best setup was evaluated by comparing the obtained 3D reconstructions with it. Once the setup has been determined, the anatomies of five volunteers were acquired, to test the methodology on human subjects.

Keywords: 3D model , Ear , Microtia , Photogrammetry , RealSense D415

The paper describes the conceptual model of an emotion-aware car interface able to: map both the driver's cognitive and emotional states with the vehicle dynamics; adapt the level of automation or support the decision-making process if emotions negatively affecting the driving performance are detected; ensure emotion regulation and provide a unique user experience creating a more engaging atmosphere (e.g. music, LED lighting) in the car cabin. To enable emotion detection, it implements a low-cost emotion recognition able to recognize Ekman's universal emotions by analyzing the driver's facial expression from stream video. A preliminary test was conducted in order to determine the effectiveness of the proposed emotion recognition system in a driving context. Results evidenced that the proposed system is capable to correctly qualify the drivers' emotion in a driving simulation context.

Keywords: Driver Monitoring System , Emotion recognition , Facial expression recognition

Recently a wide variety of applications has been developed integrating 3D functionalities. Advantages given by the possibility of relying on depth information allows the developers to design new algorithms and to improve the existing ones. In particular, for what concerns face morphology, 3D has led to the possibility to obtain face depth maps highly close to reality and consequently an improvement of the starting point for further analysis such as Face Detection, Face Authentication, Face Identification and Face Expression Recognition. The development of the aforementioned applications would have been impossible without the progress of sensor technologies for obtaining 3D information. Several solutions have been adopted over time. In this paper, emphasis is put on passive stereoscopy, structured light, time-of-flight (ToF) and active stereoscopy, namely the most used technologies for the cameras design and fulfilment according to the literature. The aim of this article is to investigate facial applications and to examine 3D camera technologies to suggest some guidelines for addressing the correct choice of a 3D sensor according to the application that has to be developed.

Keywords: 3D cameras , 3D face analysis , Sensors , Stereoscopy , Structured light , ToF

Common sense usually considers the assessment of female human attractiveness to be subjective. Nevertheless, in the past decades, several studies and experiments showed that an objective component in beauty assessment exists and can be strictly related, even if it does not match, with proportions of features. Proportions can be studied through analysis of the face, which relies on landmarks, i.e., specific points on the facial surface, which are shared by everyone, and measurements between them. In this work, several measures have been gathered from studies in the literature considering datasets of beautiful women to build a set of measures that can be defined as suggestive of female attractiveness. The resulting set consists of 29 measures applied to a public dataset, the Bosphorus database, whose faces have been both analyzed by the developed methodology based on the expanded set of measures and judged by human observers. Results show that the set of chosen measures is significant in terms of attractiveness evaluation, confirming the key role of proportions in beauty assessment; furthermore, the sorting of identified measures has been performed to identify the most significant canons involved in the evaluation.

Keywords: 3D landmarks , Attractiveness , Face analysis , Face proportions , Features extraction

Due to its versatility, virtual technology is being widely employed in different domains, from industry to amusement. The possibility to adopt this technology in early product/service design is going to bring positive effects such as the reduction of costs associated with the production of physical prototypes and the generation of a more effective knowledge of users' feedback. This study proposes a preliminary methodology to evaluate users' engagement in interacting with a virtual environment that consists of the integration between a self-report method (the user engagement scale questionnaire) and a method based on facial expression recognition. Results reported in this paper show that the two methodologies generate different types of knowledge which can be used to fit users' needs and expectations. Even if focused on a specific case study, i.e., the evaluation of the engagement in a virtual learning environment, this paper aims to propose a methodology that can be applied to all kinds of virtual products.

Keywords: FER (facial expression recognition) , Product life cycle , Self-reports , Student engagement , Virtual learning environment

Three dimensional face analysis is being widely investigated since it appears as a robust solution to overcome the limits of two dimensional technologies. 3D methods allow to relate the recognition process on features not depending on lightning, head poses, make up and occlusions. This paper proposes a new approach to the problem consisting of a novel image representation, where specific facial descriptors replace the RGB traditional channels and a convolutional neural network performs the classification. We chose to use MobileNetV2, a relatively new network, as it has a low amount of parameters to train. The method has been evaluated on the Bosphorus database, and even though it is still a preliminary study, the results obtained with our method are extremely encouraging; the recognition rate achieved is 97.560% and it is comparable to the state of the art. This result, reached despite the fact that the Bosphorus database has a great number of subjects (105) but a low number of scans (4666), shows the effectiveness of this representation combined with convolutional neural networks.

Keywords: 3D face recognition , Curvedness , Deep Convolutional Neural Network , MobileNetV2 , Shape index

In pattern recognition, neural networks can be used not only for the classification task, but also for feature selection and other intermediate steps. This paper addresses the 3D face recognition problem in order to select the most meaningful geometric descriptors. At this aim, the classification results are directly integrated in a biclustering process in order to select the best leaves of a neural hierarchical tree. This tree is created by a novel neural network GH-EXIN. This approach results in a new criterion for the feature selection. This technique is applied to a database of face expressions where both traditional and novel geometric descriptors are used. The results state the importance of the curvedness novel descriptors and only of a few Euclidean distances.

Keywords: 3D face recognition , Biclustering , Clustering , Feature selection , GH-EXIN , Neural network

Advanced Driver-Assistance Systems (ADASs) are used for increasing safety in the automotive domain, yet current ADASs notably operate without taking into account drivers' states, e.g., whether she/he is emotionally apt to drive. In this paper, we first review the state-of-the-art of emotional and cognitive analysis for ADAS: We consider psychological models, the sensors needed for capturing physiological signals, and the typical algorithms used for human emotion classification. Our investigation highlights a lack of advanced Driver Monitoring Systems (DMSs) for ADASs, which could increase driving quality and security for both drivers and passengers. We then provide our view on a novel perception architecture for driver monitoring, built around the concept of Driver Complex State (DCS). DCS relies on multiple non-obtrusive sensors and Artificial Intelligence (AI) for uncovering the driver state and uses it to implement innovative Human-Machine Interface (HMI) functionalities. This concept will be implemented and validated in the recently EU-funded NextPerception project, which is briefly introduced.

Keywords: Advanced Driver-Assistance System (ADAS) , Artificial Intelligence (AI) , Driver Complex State (DCS) , Driver safety and comfort , Emotion recognition

This paper introduces a motion analysis system based on a network of common RGB cameras, which provides the measurement of various angles considered for postural assessment, in order to facilitate the evaluation of the ergonomic indices commonly used for the determination of risk of musculoskeletal disorders of operators in manufacturing workplaces. To enable the tracking of operator postures during the performed tasks, the system exploits the multi person keypoints detection library “OpenPose”. The proposed system has been validated with a real industrial case study regarding a washing machine assembly line. Results suggest how the proposed system supports ergonomists in risk assessment of musculoskeletal disorders through the OCRA index.

Keywords: Assembly line , Ergonomic assessment , Manufacturing , Motion capture , OCRA index , Posture analysis

Vitiligo vulgaris is an autoimmune disease which causes a strong reduction of the cells producing melanin, which is the main skin pigment. This results in the growth of white patches on patients' skin, which are more or less visible, depending on the skin phototype. Precise, objective and fast detection of vitiligo patches would be crucial to produce statistically relevant data on huge populations, thus giving an insight on the disease. However, few methods are available in literature. In the present paper, a semi-automatic tool based on image processing to detect facial vitiligo patches is described. The tool requires pictures to be captured under black light illumination, which enhances patches contrast with respect to healthy skin. The user is only required to roughly define the regions of interest and set a global threshold, thus, no specific image-processing skills are required. An adaptive algorithm then automatically discerns between vitiligo and healthy skin pixels. The tools also allow for a statistical data interpretation by overlapping the detected patches of all patients on a face template through an occurrence map. Preliminary results obtained on a small population of 15 patients allowed us to assess the tool's performance. Patch detection was checked by an experienced dermatologist, who confirmed the detection for all the studied patients, thus supporting the effectiveness of the proposed tool.

Keywords: Black light , Image processing , Semi-automatic vitiligo detection , Vitiligo

The aim of this study was to confirm the identities of numerous portraits attributed to the composer Vincenzo Bellini by using 3D-to-2D projection. This study also followed on from earlier research on three death masks of Bellini, the results of which had shown that the wax mask in Catania's Bellini museum best represented Bellini's face compared to the other two. This study used the aforementioned 3D wax death mask obtained through Reverse Engineering as a reference for a morphometric comparison with 14 other portraits. For each portrait, the linear 3D-to-2D transformation M was found which minimized the distance between the 2D landmarks in the picture and the projected landmarks on the 3D mask. This normalized the distances considering the scale of the portrait and the final dissimilarity score with the mask. In particular, the analytical results were able to identify two portraits which particularly resembled the 3D death mask providing future researchers with the chance to carry out historical-artistic evaluations. We were also able to develop a new tool – Image Mark Pro - to easily annotate 2D images by introducing landmark locations. Since it was so reliable for manually annotating landmarks, we decided to make it publicly available for future research.

Keywords: 3D death mask , 3D-2D comparison , Face recognition , Landmarks projection