1,721,116 research outputs found
Fitting a biomechanical model of the folds to high-speed video data through bayesian estimation
Full text linkHigh-speed video recording of the vocal folds during sustained phonation has become a widespread diagnostic tool, and the development of imaging techniques able to perform automated tracking and analysis of relevant glottal cues, such as folds edge position or glottal area, is an active research field. In this paper, a vocal folds vibration analysis method based on the processing of visual data through a biomechanical model of the layngeal dynamics is proposed. The procedure relies on a Bayesian non-stationary estimation of the biomechanical model parameters and state, to fit the folds edge position extracted from the high-speed video endoscopic data. This finely tuned dynamical model is then used as a state transition model in a Bayesian setting, and it allows to obtain a physiologically motivated estimation of upper and lower vocal folds edge position. Based on model prediction, an hypothesis on the lower fold position can be made even in complete fold occlusion conditions occurring during the end of the closed phase and the beginning of the open phase of the glottal cycle. To demonstrate the suitability of the procedure, the method is assessed on a set of audiovisual recordings featuring high-speed video endoscopic data from healthy subjects producing sustained voiced phonation with different laryngeal settings
A hierarchical approach to feature extraction and grouping
No full textIn this paper, the problem of extracting and grouping image features from complex scenes is solved by a hierarchical approach based on two main processes: voting and clustering. Voting is performed for assigning a score to both global and local features. The score represents the evidential support provided by input data for the presence of a feature. Clustering aims at individuating a minimal set of significant local features by grouping together simpler correlated observations, It is based on a spatial relation between simple observations on a fixed level, i.e., the definition of a distance in an appropriate space. As the multilevel structure of the system implies that input data for an intermediate level are outputs of the lower level, voting can be seen as a functional representation of the "part-of" relation between features at different abstraction levels. The proposed approach has been tested on both synthetic and real images and compared with other existing feature grouping methods
A Shallow System Prototype for Violent Action Detection in Italian Public Schools
Full text linkThis paper presents a novel low-cost integrated system prototype, called School Violence Detection system (SVD), based on a 2D Convolutional Neural Network (CNN). It is used for classifying and identifying automatically violent actions in educational environments based on shallow cost hardware. Moreover, the paper fills the gap of real datasets in educational environments by proposing a new one, called Daily School Break dataset (DSB), containing original videos recorded in an Italian high school yard. The proposed CNN has been pre-trained with an ImageNet model and a transfer learning approach. To extend its capabilities, the DSB was enriched with online images representing students in school environments. Experimental results analyze the classification performances of the SVD and investigate how it performs through the proposed DSB dataset. The SVD, which achieves a recognition accuracy of 95%, is considered computably efficient and low-cost. It could be adapted to other scenarios such as school arenas, gyms, playgrounds, etc
An integrated low-cost system for object detection in underwater environments
Full text linkWe propose a novel low-cost integrated system prototype able to recognize objects/lifeforms in underwater environments. The system has been applied to detect unexploded ordnance materials in shallow waters. Indeed, small and agile remotely controlled vehicles with cameras can be used to detect unexploded bombs in shallow waters, more effectively and freely than complex, costly and heavy equipment, requiring several human operators and support boats. Moreover, visual techniques can be easily combined with the traditional use of magnetometers and scanning imaging sonars, to improve the effectiveness of the survey. The proposed system can be easily adapted to other scenarios (e.g., underwater archeology or visual inspection of underwater pipelines and implants), by simply replacing the Convolutional Neural Network devoted to the visual identification task. As a final outcome of our work we provide a large dataset of images of explosive materials: it can be used to compare different visual techniques on a common basis
A Sentiment Analysis Anomaly Detection System for Cyber Intelligence
No full textConsidering the 2030 United Nations intent of world connection, Cyber Intelligence becomes the main area of the human dimension able of inflicting changes in geopolitical dynamics. In cyberspace, the new battlefield is the mind of people including new weapons like abuse of social media with information manipulation, deception by activists and misinformation. In this paper, a Sentiment Analysis system with Anomaly Detection (SAAD) capability is proposed. The system, scalable and modular, uses an OSINT-Deep Learning approach to investigate on social media sentiment in order to predict suspicious anomaly trend in Twitter posts. Anomaly detection is investigated with a new semi-supervised process that is able to detect potentially dangerous situations in critical areas. The main contributions of the paper are the system suitability for working in different areas and domains, the anomaly detection procedure in sentiment context and a time-dependent confusion matrix to address model evaluation with unbalanced dataset. Real experiments and tests were performed on Sahel Region. The detected anomalies in negative sentiment have been checked by experts of Sahel area, proving true links between the models results and real situations observable from the tweets
An ADAS design based on IoT V2X communications to improve safety: Case study and iot architecture reference model
Full text linkA Neural Network for Image Anomaly Detection with Deep Pyramidal Representations and Dynamic Routing
Full text linkImage anomaly detection is an application-driven problem where the aim is to identify novel samples, which differ significantly from the normal ones. We here propose Pyramidal Image Anomaly DEtector (PIADE), a deep reconstruction-based pyramidal approach, in which image features are extracted at different scale levels to better catch the peculiarities that could help to discriminate between normal and anomalous data. The features are dynamically routed to a reconstruction layer and anomalies can be identified by comparing the input image with its reconstruction. Unlike similar approaches, the comparison is done by using structural similarity and perceptual loss rather than trivial pixel-by-pixel comparison. The proposed method performed at par or better than the state-of-the-art methods when tested on publicly available datasets such as CIFAR10, COIL-100 and MVTec
Editorial for the special issue on the DAFNE project (DigitalAnastylosis of Frescoes challeNgE)
No full text- …
