Alessandro Bruno - Profile on Academia.edu (original) (raw)
Papers by Alessandro Bruno
Cultural heritage understanding and preservation is an important issue for society as it represen... more Cultural heritage understanding and preservation is an important issue for society as it represents a fundamental aspect of its identity. Paintings represent a significant part of cultural heritage, and are the subject of study continuously. However, the way viewers perceive paintings is strictly related to the so-called HVS (Human Vision System) behaviour. This paper focuses on the eye-movement analysis of viewers during the visual experience of a certain number of paintings. In further details, we introduce a new approach to predicting human visual attention, which impacts several cognitive functions for humans, including the fundamental understanding of a scene, and then extend it to painting images. The proposed new architecture ingests images and returns scanpaths, a sequence of points featuring a high likelihood of catching viewers' attention. We use an FCNN (Fully Convolutional Neural Network), in which we exploit a differentiable channel-wise selection and Soft-Argmax modules. We also incorporate learnable Gaussian distributions onto the network bottleneck to simulate visual attention process bias in natural scene images. Furthermore, to reduce the effect of shifts between different domains (i.e. natural images, painting), we urge the model to learn unsupervised general features from other domains using a gradient reversal classifier. The results obtained by our model outperform existing state-of-the-art ones in terms of accuracy and efficiency.
Journal of Imaging
Liveness detection for fingerprint impressions plays a role in the meaningful prevention of any u... more Liveness detection for fingerprint impressions plays a role in the meaningful prevention of any unauthorized activity or phishing attempt. The accessibility of unique individual identification has increased the popularity of biometrics. Deep learning with computer vision has proven remarkable results in image classification, detection, and many others. The proposed methodology relies on an attention model and ResNet convolutions. Spatial attention (SA) and channel attention (CA) models were used sequentially to enhance feature learning. A three-fold sequential attention model is used along with five convolution learning layers. The method’s performances have been tested across different pooling strategies, such as Max, Average, and Stochastic, over the LivDet-2021 dataset. Comparisons against different state-of-the-art variants of Convolutional Neural Networks, such as DenseNet121, VGG19, InceptionV3, and conventional ResNet50, have been carried out. In particular, tests have been a...
Algorithms
Security in the blockchain has become a topic of concern because of the recent developments in th... more Security in the blockchain has become a topic of concern because of the recent developments in the field. One of the most common cyberattacks is the so-called phishing attack, wherein the attacker tricks the miner into adding a malicious block to the chain under genuine conditions to avoid detection and potentially destroy the entire blockchain. The current attempts at detection include the consensus protocol; however, it fails when a genuine miner tries to add a new block to the blockchain. Zero-trust policies have started making the rounds in the field as they ensure the complete detection of phishing attempts; however, they are still in the process of deployment, which may take a significant amount of time. A more accurate measure of phishing detection involves machine-learning models that use specific features to automate the entire process of classifying an attempt as either a phishing attempt or a safe attempt. This paper highlights several models that may give safe results an...
Lecture Notes in Computer Science, 2022
The advent of deep learning has brought in disruptive techniques with unprecedented accuracy rate... more The advent of deep learning has brought in disruptive techniques with unprecedented accuracy rates in so many fields and scenarios. Tasks such as the detection of regions of interest and semantic features out of images and video sequences are quite effectively tackled because of the availability of publicly available and adequately annotated datasets. This paper describes a use case scenario with a deep learning models' stack being used for crowd behaviour analysis. It consists of two main modules preceded by a pre-processing step. The first deep learning module relies on the integration of YOLOv5 and DeepSORT to detect and track down pedestrians from CCTV cameras' video sequences. The second module ingests each pedestrian's spatial coordinates, velocity, and trajectories to cluster groups of people using the Coherent Neighbor Invariance technique. The method envisages the acquisition of video sequences from cameras overlooking pedestrian areas, such as public parks or squares, in order to check out any possible unusualness in crowd behaviour. Due to its design, the system first checks whether some anomalies are underway at the microscale level. Secondly, It returns clusters of people at the mesoscale level depending on velocity and trajectories. This work is part of the physical behaviour detection module developed for the S4AllCities H2020 project.
Annals of Biomedical Engineering, 2022
Immunohistochemistry for vascular network analysis plays a fundamental role in basic science, tra... more Immunohistochemistry for vascular network analysis plays a fundamental role in basic science, translational research and clinical practice. However, identifying vascularization in histological tissue images is time consuming and markedly depends on the operator’s experience. In this study, we present “blood vessel detection—BVD”, an automatic algorithm for quantitative analysis of blood vessels in immunohistochemical images. BVD is based on extraction and analysis of low-level image features and spatial filtering techniques, which do not require a training phase. BVD algorithm performance was comparatively evaluated on histological sections from three different in vivo experiments. Collectively, 173 independent images were analyzed, and the algorithm's results were compared to those obtained by human operators. The developed BVD algorithm proved to be a robust and versatile tool, being able to quantify number, area, and spatial distribution of blood vessels within all three cons...
Proceedings of 37th International Cosmic Ray Conference — PoS(ICRC2021), 2021
Mini-EUSO launch and first results Mini-EUSO is a telescope launched on board the International S... more Mini-EUSO launch and first results Mini-EUSO is a telescope launched on board the International Space Station in 2019 and currently located in the Russian section of the station. Main scientific objectives of the mission are the search for nuclearites and Strange Quark Matter, the study of atmospheric phenomena such as Transient Luminous Events, meteors and meteoroids, the observation of sea bioluminescence and of artificial satellites and man-made space debris. It is also capable of observing Extensive Air Showers generated by Ultra-High Energy Cosmic Rays with an energy above 10 21 eV and detect artificial showers generated with lasers from the ground. Mini-EUSO can map the night-time Earth in the UV range (290 -430 nm), with a spatial resolution of about 6.3 km and a temporal resolution of 2.5 s, observing our planet through a nadir-facing UV-transparent window in the Russian Zvezda module. The instrument, launched on 2019/08/22 from the Baikonur cosmodrome, is based on an optical system employing two Fresnel lenses and a focal surface composed of 36 Multi-Anode Photomultiplier tubes, 64 channels each, for a total of 2304 channels with single photon counting sensitivity and an overall field of view of 44 • . Mini-EUSO also contains two ancillary cameras to complement measurements in the near infrared and visible ranges. In this paper we describe the detector and present the various phenomena observed in the first year of operation.
Advances in Neuroergonomics and Cognitive Engineering, 2021
The ever-growing adoption of big data technologies, smart sensing, data science and artificial in... more The ever-growing adoption of big data technologies, smart sensing, data science and artificial intelligence is enabling the development of new intelligent urban spaces with real-time monitoring and advanced cyber-physical situational awareness capabilities. In the S4AllCities international research project, the advancement of cyber-physical situational awareness will be experimented for achieving safer smart city spaces in Europe and beyond. The deployment of digital twins will lead to understanding real-time situation awareness and risks of potential physical and/or cyber-attacks on urban critical infrastructure specifically. The critical extraction of knowledge using digital twins, which ingest, process and fuse observation data and information, prior to machine reasoning is performed in S4AllCities. In this paper, a cyber behavior detection module, which identifies unusualness in cyber traffic networks is described. Also, a physical behaviour detection module is introduced. The two modules function within the so-called Malicious Attacks Information Detection System (MAIDS) digital twin.
Proceedings of 37th International Cosmic Ray Conference — PoS(ICRC2021), 2021
Very-High Energy (VHE) gamma-ray astroparticle physics is a relatively young field, and observati... more Very-High Energy (VHE) gamma-ray astroparticle physics is a relatively young field, and observations over the past decade have surprisingly revealed almost two hundred VHE emitters which appear to act as cosmic particle accelerators. These sources are an important component of the Universe, influencing the evolution of stars and galaxies. At the same time, they also act as a probe of physics in the most extreme environments known -such as in supernova explosions, and around or after the merging of black holes and neutron stars. However, the existing experiments have provided exciting glimpses, but often falling short of supplying the full answer. A deeper understanding of the TeV sky requires a significant improvement in sensitivity at TeV energies, a wider energy coverage from tens of GeV to hundreds of TeV and a much better angular and energy resolution with respect to the currently running facilities. The next generation gamma-ray observatory, the Cherenkov Telescope Array Observatory (CTAO), is the answer to this need. In this talk I will present this upcoming observatory from its design to the construction, and its potential science exploitation. CTAO will allow the entire astronomical community to explore a new discovery space that will likely lead to paradigm-changing breakthroughs. In particular, CTA has an unprecedented sensitivity to short (sub-minute) timescale phenomena, placing it as a key instrument in the future of multi-messenger and multi-wavelength time domain astronomy. I will conclude the talk presenting the first scientific results obtained by the LST-1, the prototype of one CTA telescope type -the Large Sized Telescope, that is currently under commission.
arXiv (Cornell University), Jul 10, 2023
Knee osteoarthritis (KOA) is a widespread condition that can cause chronic pain and stiffness in ... more Knee osteoarthritis (KOA) is a widespread condition that can cause chronic pain and stiffness in the knee joint. Early detection and diagnosis are crucial for successful clinical intervention and management to prevent severe complications, such as loss of mobility. In this paper, we propose an automated approach that employs the Swin Transformer to predict the severity of KOA. Our model uses publicly available radiographic datasets with Kellgren and Lawrence scores to enable early detection and severity assessment. To improve the accuracy of our model, we employ a multi-prediction head architecture that utilizes multi-layer perceptron classifiers. Additionally, we introduce a novel training approach that reduces the data drift between multiple datasets to ensure the generalization ability of the model. The results of our experiments demonstrate the effectiveness and feasibility of our approach in predicting KOA severity accurately.
Journal of Imaging
Research in the medical imaging field using deep learning approaches has become progressively con... more Research in the medical imaging field using deep learning approaches has become progressively contingent. Scientific findings reveal that supervised deep learning methods’ performance heavily depends on training set size, which expert radiologists must manually annotate. The latter is quite a tiring and time-consuming task. Therefore, most of the freely accessible biomedical image datasets are small-sized. Furthermore, it is challenging to have big-sized medical image datasets due to privacy and legal issues. Consequently, not a small number of supervised deep learning models are prone to overfitting and cannot produce generalized output. One of the most popular methods to mitigate the issue above goes under the name of data augmentation. This technique helps increase training set size by utilizing various transformations and has been publicized to improve the model performance when tested on new data. This article surveyed different data augmentation techniques employed on mammogra...
Journal of Imaging
Human beings usually rely on communication to express their feeling and ideas and to solve disput... more Human beings usually rely on communication to express their feeling and ideas and to solve disputes among themselves. A major component required for effective communication is language. Language can occur in different forms, including written symbols, gestures, and vocalizations. It is usually essential for all of the communicating parties to be fully conversant with a common language. However, to date this has not been the case between speech-impaired people who use sign language and people who use spoken languages. A number of different studies have pointed out a significant gaps between these two groups which can limit the ease of communication. Therefore, this study aims to develop an efficient deep learning model that can be used to predict British sign language in an attempt to narrow this communication gap between speech-impaired and non-speech-impaired people in the community. Two models were developed in this research, CNN and LSTM, and their performance was evaluated using...
Proceedings of the 7th International Conference on Sensor Networks, 2018
United States National Oceanic and Atmospheric Administration (NOAA) weather satellites adopt Adv... more United States National Oceanic and Atmospheric Administration (NOAA) weather satellites adopt Advanced Very High Resolution Radiometer (AVHRR) sensors to acquire remote sensing data and broadcast Automatic Picture Transmission (APT) images. The orientation of the scan lines is perpendicular to the orbit of the satellite. In this paper we propose a new low cost solution for NOAA remote sensing. More in detail, our method focuses on the possibility of directly sampling the modulated signal and processing it entirely in software enabled by recent breakthroughs on Software Defined Radios (SDR) and CPU computational speed, while keeping the costs extremely low. We aim to achieve good results with inexpensive SDR hardware, like the RTL-SDR (a repurposed DVB-T USB dongle). Nevertheless, we faced some problems caused by hardware limits such as high receiver noise figure and low ADC resolution. Furthermore, we detected several inherent drawbacks of frequent tuner saturations. For this purpose we developed a software-hardware integrated system able to perform the following steps: satellite pass prediction, time scheduling, signal demodulation, image cropping and filtering. Although we employed low cost components, we obtained good results in terms of signal demodulation, synchronization and image reconstruction.
Proceedings of the 3rd International Conference on Pattern Recognition Applications and Methods, 2014
In this paper we present a novel technique for object modeling and object recognition in video. G... more In this paper we present a novel technique for object modeling and object recognition in video. Given a set of videos containing 360 degrees views of objects we compute a model for each object, then we analyze short videos to determine if the object depicted in the video is one of the modeled objects. The object model is built from a video spanning a 360 degree view of the object taken against a uniform background. In order to create the object model, the proposed techniques selects a few representative frames from each video and local features of such frames. The object recognition is performed selecting a few frames from the query video, extracting local features from each frame and looking for matches in all the representative frames constituting the models of all the objects. If the number of matches exceed a fixed threshold the corresponding object is considered the recognized objects .To evaluate our approach we acquired a dataset of 25 videos representing 25 different objects and used these videos to build the objects model. Then we took 25 test videos containing only one of the known objects and 5 videos containing only unknown objects. Experiments showed that, despite a significant compression in the model, recognition results are satisfactory.
Soft Matter, 2018
Fiber intersection density affects meso-scale cell aspect ratio and extracellular matrix synthesi... more Fiber intersection density affects meso-scale cell aspect ratio and extracellular matrix synthesis in an elastomeric scaffold model under organ-scale deformation.
IEEE Access, 2022
Human visual Attention modelling is a persistent interdisciplinary research challenge, gaining ne... more Human visual Attention modelling is a persistent interdisciplinary research challenge, gaining new interest in recent years mainly due to the latest developments in deep learning. That is particularly evident in saliency benchmarks. Novel deep learning-based visual saliency models show promising results in capturing high-level (top-down) human visual attention processes. Therefore, they strongly differ from the earlier approaches, mainly characterised by low-level (bottom-up) visual features. These developments account for innate human selectivity mechanisms that are reliant on both high-and low-level factors. Moreover, the two factors interact with each other. Motivated by the importance of these interactions, in this project, we tackle visual saliency modelling holistically, examining if we could consider both high-and low-level features that govern human attention. Specifically, we propose a novel method SAtSal (Self-Attention Saliency). SAtSal leverages both high and low-level features using a multilevel merging of skip connections during the decoding stage. Consequently, we incorporate convolutional self-attention modules on skip connection from the encoder to the decoder network to properly integrate the valuable signals from multilevel spatial features. Thus, the self-attention modules learn to filter out the latent representation of the salient regions from the other irrelevant information in an embedded and joint manner with the main encoder-decoder model backbone. Finally, we evaluate SAtSal against various existing solutions to validate our approach, using the well-known standard saliency benchmark MIT300. To further examine SAtSal's robustness on other image types, we also evaluate it on the Le-Meur saliency painting benchmark. INDEX TERMS Eye movements, low and high vision, saliency prediction, self-attention, visual attention.
IEEE Access, 2020
In the first seconds of observation of an image, several visual attention processes are involved ... more In the first seconds of observation of an image, several visual attention processes are involved in the identification of the visual targets that pop-out from the scene to our eyes. Saliency is the quality that makes certain regions of an image stand out from the visual field and grab our attention. Saliency detection models, inspired by visual cortex mechanisms, employ both colour and luminance features. Furthermore, both locations of pixels and presence of objects influence the Visual Attention processes. In this paper, we propose a new saliency method based on the combination of the distribution of interest points in the image with multiscale analysis, a centre bias module and a machine learning approach. We use perceptually uniform colour spaces to study how colour impacts on the extraction of saliency. To investigate eye-movements and assess the performances of saliency methods over object-based images, we conduct experimental sessions on our dataset ETTO (Eye Tracking Through Objects). Experiments show our approach to be accurate in the detection of saliency concerning state-of-the-art methods and accessible eye-movement datasets. The performances over object-based images are excellent and consistent on generic pictures. Besides, our work reveals interesting findings on some relationships between saliency and perceptually uniform colour spaces. INDEX TERMS Eye-movements, interest points, saliency map, visual attention.
Sul significato delle costanti logiche: Paradosso dell'inferenza e teoria dei giochi Massimo Panz... more Sul significato delle costanti logiche: Paradosso dell'inferenza e teoria dei giochi Massimo Panzarella Un'analisi preliminare della rete dei ringraziamenti su Wikipedia Valerio Perticone, Marco Elio Tabacchi Alla ricerca del circuito per l'intelligenza Alessio Plebe L'Opinion Mining nelle Scienze Cognitive: espressione dei sentimenti e reti sociali
Image Analysis and Processing - ICIAP 2017, 2017
In this paper an automatic multi-seed detection method for magnetic resonance (MR) breast image s... more In this paper an automatic multi-seed detection method for magnetic resonance (MR) breast image segmentation is presented. The proposed method consists of three steps: (1) pre-processing step to locate three regions of interest (axillary and sternal regions); (2) processing step to detect maximum concavity points for each region of interest; (3) breast image segmentation step. Traditional manual segmentation methods require radiological expertise and they usually are very tiring and time-consuming. The approach is fast because the multi-seed detection is based on geometric properties of the ROI. When the maximum concavity points of the breast regions have been detected, region growing and morphological transforms complete the segmentation of breast MR image. In order to create a Gold Standard for method effectiveness and comparison, a dataset composed of 18 patients is selected, accordingly to three expert radiologists of University of Palermo Policlinico Hospital (UPPH). Each patient has been manually segmented. The proposed method shows very encouraging results in terms of statistical metrics
Image Analysis and Processing - ICIAP 2017, 2017
Visual Saliency aims to detect the most important regions of an image from a perceptual point of ... more Visual Saliency aims to detect the most important regions of an image from a perceptual point of view. More in detail, the goal of Visual Saliency is to build a Saliency Map revealing the salient subset of a given image by analyzing bottom-up and top-down factors of Visual Attention. In this paper we proposed a new method for Saliency detection based on colour and scale analysis, extending our previous work based on SIFT spatial density inspection. We conducted several experiments to study the relationships between saliency methods and the object attention processes and we collected experimental data by tracking the eye movements of thirty viewers in the first three seconds of observation of several images. More precisely, we used a dataset that consists of images with an object in the foreground on an homogeneous background. We are interested in studying the performance of our saliency method with respect to the real fixation maps collected during the experiments. We compared the performances of our method with several state of the art methods with very encouraging results.
Cultural heritage understanding and preservation is an important issue for society as it represen... more Cultural heritage understanding and preservation is an important issue for society as it represents a fundamental aspect of its identity. Paintings represent a significant part of cultural heritage, and are the subject of study continuously. However, the way viewers perceive paintings is strictly related to the so-called HVS (Human Vision System) behaviour. This paper focuses on the eye-movement analysis of viewers during the visual experience of a certain number of paintings. In further details, we introduce a new approach to predicting human visual attention, which impacts several cognitive functions for humans, including the fundamental understanding of a scene, and then extend it to painting images. The proposed new architecture ingests images and returns scanpaths, a sequence of points featuring a high likelihood of catching viewers' attention. We use an FCNN (Fully Convolutional Neural Network), in which we exploit a differentiable channel-wise selection and Soft-Argmax modules. We also incorporate learnable Gaussian distributions onto the network bottleneck to simulate visual attention process bias in natural scene images. Furthermore, to reduce the effect of shifts between different domains (i.e. natural images, painting), we urge the model to learn unsupervised general features from other domains using a gradient reversal classifier. The results obtained by our model outperform existing state-of-the-art ones in terms of accuracy and efficiency.
Journal of Imaging
Liveness detection for fingerprint impressions plays a role in the meaningful prevention of any u... more Liveness detection for fingerprint impressions plays a role in the meaningful prevention of any unauthorized activity or phishing attempt. The accessibility of unique individual identification has increased the popularity of biometrics. Deep learning with computer vision has proven remarkable results in image classification, detection, and many others. The proposed methodology relies on an attention model and ResNet convolutions. Spatial attention (SA) and channel attention (CA) models were used sequentially to enhance feature learning. A three-fold sequential attention model is used along with five convolution learning layers. The method’s performances have been tested across different pooling strategies, such as Max, Average, and Stochastic, over the LivDet-2021 dataset. Comparisons against different state-of-the-art variants of Convolutional Neural Networks, such as DenseNet121, VGG19, InceptionV3, and conventional ResNet50, have been carried out. In particular, tests have been a...
Algorithms
Security in the blockchain has become a topic of concern because of the recent developments in th... more Security in the blockchain has become a topic of concern because of the recent developments in the field. One of the most common cyberattacks is the so-called phishing attack, wherein the attacker tricks the miner into adding a malicious block to the chain under genuine conditions to avoid detection and potentially destroy the entire blockchain. The current attempts at detection include the consensus protocol; however, it fails when a genuine miner tries to add a new block to the blockchain. Zero-trust policies have started making the rounds in the field as they ensure the complete detection of phishing attempts; however, they are still in the process of deployment, which may take a significant amount of time. A more accurate measure of phishing detection involves machine-learning models that use specific features to automate the entire process of classifying an attempt as either a phishing attempt or a safe attempt. This paper highlights several models that may give safe results an...
Lecture Notes in Computer Science, 2022
The advent of deep learning has brought in disruptive techniques with unprecedented accuracy rate... more The advent of deep learning has brought in disruptive techniques with unprecedented accuracy rates in so many fields and scenarios. Tasks such as the detection of regions of interest and semantic features out of images and video sequences are quite effectively tackled because of the availability of publicly available and adequately annotated datasets. This paper describes a use case scenario with a deep learning models' stack being used for crowd behaviour analysis. It consists of two main modules preceded by a pre-processing step. The first deep learning module relies on the integration of YOLOv5 and DeepSORT to detect and track down pedestrians from CCTV cameras' video sequences. The second module ingests each pedestrian's spatial coordinates, velocity, and trajectories to cluster groups of people using the Coherent Neighbor Invariance technique. The method envisages the acquisition of video sequences from cameras overlooking pedestrian areas, such as public parks or squares, in order to check out any possible unusualness in crowd behaviour. Due to its design, the system first checks whether some anomalies are underway at the microscale level. Secondly, It returns clusters of people at the mesoscale level depending on velocity and trajectories. This work is part of the physical behaviour detection module developed for the S4AllCities H2020 project.
Annals of Biomedical Engineering, 2022
Immunohistochemistry for vascular network analysis plays a fundamental role in basic science, tra... more Immunohistochemistry for vascular network analysis plays a fundamental role in basic science, translational research and clinical practice. However, identifying vascularization in histological tissue images is time consuming and markedly depends on the operator’s experience. In this study, we present “blood vessel detection—BVD”, an automatic algorithm for quantitative analysis of blood vessels in immunohistochemical images. BVD is based on extraction and analysis of low-level image features and spatial filtering techniques, which do not require a training phase. BVD algorithm performance was comparatively evaluated on histological sections from three different in vivo experiments. Collectively, 173 independent images were analyzed, and the algorithm's results were compared to those obtained by human operators. The developed BVD algorithm proved to be a robust and versatile tool, being able to quantify number, area, and spatial distribution of blood vessels within all three cons...
Proceedings of 37th International Cosmic Ray Conference — PoS(ICRC2021), 2021
Mini-EUSO launch and first results Mini-EUSO is a telescope launched on board the International S... more Mini-EUSO launch and first results Mini-EUSO is a telescope launched on board the International Space Station in 2019 and currently located in the Russian section of the station. Main scientific objectives of the mission are the search for nuclearites and Strange Quark Matter, the study of atmospheric phenomena such as Transient Luminous Events, meteors and meteoroids, the observation of sea bioluminescence and of artificial satellites and man-made space debris. It is also capable of observing Extensive Air Showers generated by Ultra-High Energy Cosmic Rays with an energy above 10 21 eV and detect artificial showers generated with lasers from the ground. Mini-EUSO can map the night-time Earth in the UV range (290 -430 nm), with a spatial resolution of about 6.3 km and a temporal resolution of 2.5 s, observing our planet through a nadir-facing UV-transparent window in the Russian Zvezda module. The instrument, launched on 2019/08/22 from the Baikonur cosmodrome, is based on an optical system employing two Fresnel lenses and a focal surface composed of 36 Multi-Anode Photomultiplier tubes, 64 channels each, for a total of 2304 channels with single photon counting sensitivity and an overall field of view of 44 • . Mini-EUSO also contains two ancillary cameras to complement measurements in the near infrared and visible ranges. In this paper we describe the detector and present the various phenomena observed in the first year of operation.
Advances in Neuroergonomics and Cognitive Engineering, 2021
The ever-growing adoption of big data technologies, smart sensing, data science and artificial in... more The ever-growing adoption of big data technologies, smart sensing, data science and artificial intelligence is enabling the development of new intelligent urban spaces with real-time monitoring and advanced cyber-physical situational awareness capabilities. In the S4AllCities international research project, the advancement of cyber-physical situational awareness will be experimented for achieving safer smart city spaces in Europe and beyond. The deployment of digital twins will lead to understanding real-time situation awareness and risks of potential physical and/or cyber-attacks on urban critical infrastructure specifically. The critical extraction of knowledge using digital twins, which ingest, process and fuse observation data and information, prior to machine reasoning is performed in S4AllCities. In this paper, a cyber behavior detection module, which identifies unusualness in cyber traffic networks is described. Also, a physical behaviour detection module is introduced. The two modules function within the so-called Malicious Attacks Information Detection System (MAIDS) digital twin.
Proceedings of 37th International Cosmic Ray Conference — PoS(ICRC2021), 2021
Very-High Energy (VHE) gamma-ray astroparticle physics is a relatively young field, and observati... more Very-High Energy (VHE) gamma-ray astroparticle physics is a relatively young field, and observations over the past decade have surprisingly revealed almost two hundred VHE emitters which appear to act as cosmic particle accelerators. These sources are an important component of the Universe, influencing the evolution of stars and galaxies. At the same time, they also act as a probe of physics in the most extreme environments known -such as in supernova explosions, and around or after the merging of black holes and neutron stars. However, the existing experiments have provided exciting glimpses, but often falling short of supplying the full answer. A deeper understanding of the TeV sky requires a significant improvement in sensitivity at TeV energies, a wider energy coverage from tens of GeV to hundreds of TeV and a much better angular and energy resolution with respect to the currently running facilities. The next generation gamma-ray observatory, the Cherenkov Telescope Array Observatory (CTAO), is the answer to this need. In this talk I will present this upcoming observatory from its design to the construction, and its potential science exploitation. CTAO will allow the entire astronomical community to explore a new discovery space that will likely lead to paradigm-changing breakthroughs. In particular, CTA has an unprecedented sensitivity to short (sub-minute) timescale phenomena, placing it as a key instrument in the future of multi-messenger and multi-wavelength time domain astronomy. I will conclude the talk presenting the first scientific results obtained by the LST-1, the prototype of one CTA telescope type -the Large Sized Telescope, that is currently under commission.
arXiv (Cornell University), Jul 10, 2023
Knee osteoarthritis (KOA) is a widespread condition that can cause chronic pain and stiffness in ... more Knee osteoarthritis (KOA) is a widespread condition that can cause chronic pain and stiffness in the knee joint. Early detection and diagnosis are crucial for successful clinical intervention and management to prevent severe complications, such as loss of mobility. In this paper, we propose an automated approach that employs the Swin Transformer to predict the severity of KOA. Our model uses publicly available radiographic datasets with Kellgren and Lawrence scores to enable early detection and severity assessment. To improve the accuracy of our model, we employ a multi-prediction head architecture that utilizes multi-layer perceptron classifiers. Additionally, we introduce a novel training approach that reduces the data drift between multiple datasets to ensure the generalization ability of the model. The results of our experiments demonstrate the effectiveness and feasibility of our approach in predicting KOA severity accurately.
Journal of Imaging
Research in the medical imaging field using deep learning approaches has become progressively con... more Research in the medical imaging field using deep learning approaches has become progressively contingent. Scientific findings reveal that supervised deep learning methods’ performance heavily depends on training set size, which expert radiologists must manually annotate. The latter is quite a tiring and time-consuming task. Therefore, most of the freely accessible biomedical image datasets are small-sized. Furthermore, it is challenging to have big-sized medical image datasets due to privacy and legal issues. Consequently, not a small number of supervised deep learning models are prone to overfitting and cannot produce generalized output. One of the most popular methods to mitigate the issue above goes under the name of data augmentation. This technique helps increase training set size by utilizing various transformations and has been publicized to improve the model performance when tested on new data. This article surveyed different data augmentation techniques employed on mammogra...
Journal of Imaging
Human beings usually rely on communication to express their feeling and ideas and to solve disput... more Human beings usually rely on communication to express their feeling and ideas and to solve disputes among themselves. A major component required for effective communication is language. Language can occur in different forms, including written symbols, gestures, and vocalizations. It is usually essential for all of the communicating parties to be fully conversant with a common language. However, to date this has not been the case between speech-impaired people who use sign language and people who use spoken languages. A number of different studies have pointed out a significant gaps between these two groups which can limit the ease of communication. Therefore, this study aims to develop an efficient deep learning model that can be used to predict British sign language in an attempt to narrow this communication gap between speech-impaired and non-speech-impaired people in the community. Two models were developed in this research, CNN and LSTM, and their performance was evaluated using...
Proceedings of the 7th International Conference on Sensor Networks, 2018
United States National Oceanic and Atmospheric Administration (NOAA) weather satellites adopt Adv... more United States National Oceanic and Atmospheric Administration (NOAA) weather satellites adopt Advanced Very High Resolution Radiometer (AVHRR) sensors to acquire remote sensing data and broadcast Automatic Picture Transmission (APT) images. The orientation of the scan lines is perpendicular to the orbit of the satellite. In this paper we propose a new low cost solution for NOAA remote sensing. More in detail, our method focuses on the possibility of directly sampling the modulated signal and processing it entirely in software enabled by recent breakthroughs on Software Defined Radios (SDR) and CPU computational speed, while keeping the costs extremely low. We aim to achieve good results with inexpensive SDR hardware, like the RTL-SDR (a repurposed DVB-T USB dongle). Nevertheless, we faced some problems caused by hardware limits such as high receiver noise figure and low ADC resolution. Furthermore, we detected several inherent drawbacks of frequent tuner saturations. For this purpose we developed a software-hardware integrated system able to perform the following steps: satellite pass prediction, time scheduling, signal demodulation, image cropping and filtering. Although we employed low cost components, we obtained good results in terms of signal demodulation, synchronization and image reconstruction.
Proceedings of the 3rd International Conference on Pattern Recognition Applications and Methods, 2014
In this paper we present a novel technique for object modeling and object recognition in video. G... more In this paper we present a novel technique for object modeling and object recognition in video. Given a set of videos containing 360 degrees views of objects we compute a model for each object, then we analyze short videos to determine if the object depicted in the video is one of the modeled objects. The object model is built from a video spanning a 360 degree view of the object taken against a uniform background. In order to create the object model, the proposed techniques selects a few representative frames from each video and local features of such frames. The object recognition is performed selecting a few frames from the query video, extracting local features from each frame and looking for matches in all the representative frames constituting the models of all the objects. If the number of matches exceed a fixed threshold the corresponding object is considered the recognized objects .To evaluate our approach we acquired a dataset of 25 videos representing 25 different objects and used these videos to build the objects model. Then we took 25 test videos containing only one of the known objects and 5 videos containing only unknown objects. Experiments showed that, despite a significant compression in the model, recognition results are satisfactory.
Soft Matter, 2018
Fiber intersection density affects meso-scale cell aspect ratio and extracellular matrix synthesi... more Fiber intersection density affects meso-scale cell aspect ratio and extracellular matrix synthesis in an elastomeric scaffold model under organ-scale deformation.
IEEE Access, 2022
Human visual Attention modelling is a persistent interdisciplinary research challenge, gaining ne... more Human visual Attention modelling is a persistent interdisciplinary research challenge, gaining new interest in recent years mainly due to the latest developments in deep learning. That is particularly evident in saliency benchmarks. Novel deep learning-based visual saliency models show promising results in capturing high-level (top-down) human visual attention processes. Therefore, they strongly differ from the earlier approaches, mainly characterised by low-level (bottom-up) visual features. These developments account for innate human selectivity mechanisms that are reliant on both high-and low-level factors. Moreover, the two factors interact with each other. Motivated by the importance of these interactions, in this project, we tackle visual saliency modelling holistically, examining if we could consider both high-and low-level features that govern human attention. Specifically, we propose a novel method SAtSal (Self-Attention Saliency). SAtSal leverages both high and low-level features using a multilevel merging of skip connections during the decoding stage. Consequently, we incorporate convolutional self-attention modules on skip connection from the encoder to the decoder network to properly integrate the valuable signals from multilevel spatial features. Thus, the self-attention modules learn to filter out the latent representation of the salient regions from the other irrelevant information in an embedded and joint manner with the main encoder-decoder model backbone. Finally, we evaluate SAtSal against various existing solutions to validate our approach, using the well-known standard saliency benchmark MIT300. To further examine SAtSal's robustness on other image types, we also evaluate it on the Le-Meur saliency painting benchmark. INDEX TERMS Eye movements, low and high vision, saliency prediction, self-attention, visual attention.
IEEE Access, 2020
In the first seconds of observation of an image, several visual attention processes are involved ... more In the first seconds of observation of an image, several visual attention processes are involved in the identification of the visual targets that pop-out from the scene to our eyes. Saliency is the quality that makes certain regions of an image stand out from the visual field and grab our attention. Saliency detection models, inspired by visual cortex mechanisms, employ both colour and luminance features. Furthermore, both locations of pixels and presence of objects influence the Visual Attention processes. In this paper, we propose a new saliency method based on the combination of the distribution of interest points in the image with multiscale analysis, a centre bias module and a machine learning approach. We use perceptually uniform colour spaces to study how colour impacts on the extraction of saliency. To investigate eye-movements and assess the performances of saliency methods over object-based images, we conduct experimental sessions on our dataset ETTO (Eye Tracking Through Objects). Experiments show our approach to be accurate in the detection of saliency concerning state-of-the-art methods and accessible eye-movement datasets. The performances over object-based images are excellent and consistent on generic pictures. Besides, our work reveals interesting findings on some relationships between saliency and perceptually uniform colour spaces. INDEX TERMS Eye-movements, interest points, saliency map, visual attention.
Sul significato delle costanti logiche: Paradosso dell'inferenza e teoria dei giochi Massimo Panz... more Sul significato delle costanti logiche: Paradosso dell'inferenza e teoria dei giochi Massimo Panzarella Un'analisi preliminare della rete dei ringraziamenti su Wikipedia Valerio Perticone, Marco Elio Tabacchi Alla ricerca del circuito per l'intelligenza Alessio Plebe L'Opinion Mining nelle Scienze Cognitive: espressione dei sentimenti e reti sociali
Image Analysis and Processing - ICIAP 2017, 2017
In this paper an automatic multi-seed detection method for magnetic resonance (MR) breast image s... more In this paper an automatic multi-seed detection method for magnetic resonance (MR) breast image segmentation is presented. The proposed method consists of three steps: (1) pre-processing step to locate three regions of interest (axillary and sternal regions); (2) processing step to detect maximum concavity points for each region of interest; (3) breast image segmentation step. Traditional manual segmentation methods require radiological expertise and they usually are very tiring and time-consuming. The approach is fast because the multi-seed detection is based on geometric properties of the ROI. When the maximum concavity points of the breast regions have been detected, region growing and morphological transforms complete the segmentation of breast MR image. In order to create a Gold Standard for method effectiveness and comparison, a dataset composed of 18 patients is selected, accordingly to three expert radiologists of University of Palermo Policlinico Hospital (UPPH). Each patient has been manually segmented. The proposed method shows very encouraging results in terms of statistical metrics
Image Analysis and Processing - ICIAP 2017, 2017
Visual Saliency aims to detect the most important regions of an image from a perceptual point of ... more Visual Saliency aims to detect the most important regions of an image from a perceptual point of view. More in detail, the goal of Visual Saliency is to build a Saliency Map revealing the salient subset of a given image by analyzing bottom-up and top-down factors of Visual Attention. In this paper we proposed a new method for Saliency detection based on colour and scale analysis, extending our previous work based on SIFT spatial density inspection. We conducted several experiments to study the relationships between saliency methods and the object attention processes and we collected experimental data by tracking the eye movements of thirty viewers in the first three seconds of observation of several images. More precisely, we used a dataset that consists of images with an object in the foreground on an homogeneous background. We are interested in studying the performance of our saliency method with respect to the real fixation maps collected during the experiments. We compared the performances of our method with several state of the art methods with very encouraging results.