1,721,154 research outputs found
A clustering fuzzy approach for image segmentation
No full textSegmentation is a fundamental step in image description or classification. In recent years, several computational models have been used to implement segmentation methods but without establishing a single analytic solution. However, the intrinsic properties of neural networks make them an interesting approach, despite some measure of inefficiency. This paper presents aclusteringapproach for imagesegmentation based on a modified fuzzyapproach for imagesegmentation (ART) model. The goal of the proposed approach is to find a simple model able to instance a prototype for each cluster avoiding complex post-processing phases. Results and comparisons with other similar models presented in the literature (like self-organizing maps and original fuzzy ART) are also discussed. Qualitative and quantitative evaluations confirm the validity of the approach proposed
L'intelligence del futuro
No full textI saggi raccolti nel presente volume sono il risultato di alcune delle ricerche e degli studi nati all'interno delle prime due edizioni del Master in Intelligence & ICT del Dipartimento di Matematica, Informatica e Fisica dell'Università degli Studi di Udine, un percorso disciplinare unico nel suo genere, che integra una preparazione mirata alla gestione consapevole delle emergenti tecnologie digitali (intelligenza artificiale, machine e deep learning, big data analysis, cyber security, biometria e gait analysis, web intelligence, text mining, open source intelligence, crowdsource intelligence, augmented e virtual reality, ecc.), con la capacità, imprescindibile per l'analista dell'intelligence di oggi, di interpretare, a fini predittivi, la realtà.
Questa capacità interpretativa non può prescindere da una formazione in grado di fornire non solo una competenza tecnica ben allenata a muoversi nell'inesauribile mole di dati prodotti dalla tecnologia, ma anche gli strumenti per organizzare e dare un senso a tali dati.
I contributi raccolti nel libro possono essere suddivisi in tre macro-ambiti che caratterizzano il percorso di studi. Il primo è quello relativo alle tecnologie emergenti, alla cyber intelligence (machine e deep learning applicate alle operazioni di cibernetiche militari, all'anomaly detection, alla predizione di comportamenti di gruppi terroristici) e alle tecnologie basate sulla blockchain. Il secondo riguarda la storia dell'intelligence e il rapporto di quest'ultima con l'evoluzione tecnologica, gli elementi generali del diritto per l'intelligence, i fondamenti di geopolitica e i fondamenti di intelligence economica. Nel terzo ambito, infine, sono raccolti contributi relativi ai fondamenti del ciclo dell'intelligence, quelli del rapporto tra comunicazione e intelligence e quelli tra open source intelligence e analisi dei fenomeni sociali
Real-Time Hand Gesture Recognition Using a Color Glove.
No full textThis paper presents a real-time hand gesture recognizer based on a color glove. The recognizer is formed by three modules. The first module, fed by the frame acquired by a webcam, identifies the hand image in the scene. The second module, a feature extractor, represents the image by a nine-dimensional feature vector. The third module, the classifier, is performed by means of Learning Vector Quantization. The recognizer, tested on a dataset of 907 hand gestures, has shown very high recognition rate
A multimodal learning system for individuals with sensorial, neuropsychological and relational impairments
Full text linkThis paper presents a system for an interactive multimodal environment able (i) to train the listening comprehension in various populations of pupils, both Italian and immigrants, having different disabilities and (ii) to assess speech production and discrimination. The proposed system is the result of a research project focused on pupils with sensorial, neuropsychological, and relational impairments. The project involves innovative technological systems that the users (speech terabits psychologists and preprimary and primary schools teachers) could adopt for training and assessment of language and speech. Because the system is used in a real scenario (the Italian schools are often affected by poor funding for education and teachers without informatics skills), the guidelines adopted are low-cost technolog
The Specchieri MarVen Dataset: an Abbreviation-Rich Dataset in Venetian Idiom
No full textDespite the release of numerous datasets for training models in historical handwritten text recognition, there is still a significant need for more diverse and extensive data. This paper aims to contribute to bridging this gap by introducing a new dataset comprising 159 pages from an Early Modern age volume part of the Venetian ‘Marigold’ collection. The dataset contains various abbreviations that are key to transcribing for a complete understanding of the content. To accommodate different research needs, the dataset is released in two versions: one with ‘expanded’ abbreviations and another without abbreviations – where the abbreviations are removed –, aligning with the choices made for other released datasets. Additionally, the dataset encompasses two distinct writing styles, leading us to provide three separate splits for training and evaluating machine learning models: one with a combination of both styles and two individual splits for each style. The qualitative and quantitative characteristics of all dataset configurations are analysed. In addition, three diverse architectures for handwritten text recognition are trained to assess their performances on this dataset. The dataset is available for download at https://doi.org/10.48557/GJYJTW
Shaping the Error-Reject Curve of Error Correcting Output Coding Systems
No full textA common approach in many classification tasks consists in reducing the costs by turning as many errors as possible into rejects. This can be accomplished by introducing a reject rule which, working on the reliability of the decision, aims at increasing the performance of the classification system. When facing multiclass classification, Error Correcting Output Coding is a diffused and successful technique to implement a system by decomposing the original problem into a set of two class problems. The novelty in this paper is to consider different levels where the reject can be applied in the ECOC systems. A study for the behavior of such rules in terms of Error-Reject curves is also proposed and tested on several benchmark datasets
Exploring Cascade Classifiers for Detecting Clusters of Microcalcifications
No full textThe conventional approach to the detection of microcalcifications on mammographies is to employ a sliding window technique. This consists in applying a classifier function to all the subwindows contained in an image and taking each local maximum of the classifier as a possible position of a microcalcification. Although effective such an approach suffers from the high computational burden due to the huge number of subwindows contained in an image. The aim of this paper is to experimentally verify if such problem can be alleviated by a detection system which employs a cascade-based localization coupled with a clustering algorithm which exploits both the spatial coordinates of the localized regions and a confidence degree estimated on them by the final stage of the cascade. The first results obtained on a publicly available set of mammograms show that the method is promising and has large
possibility of improvement
Bringing Attention to Image Anomaly Detection
No full textDetecting anomalies in images is a task with several relevant real-world applications, e.g. industrial inspection. Building on the existing RIAD (Reconstruction by Inpainting for visual Anomaly Detection) framework, we introduce an attention-based component to improve the model performance. Furthermore we propose a different approach to image masking which leverages the selection of multiple random patches at a single scale in the original images. Through the provided experimental results we show how the novelties introduced by this work consistently improve the performance of the baseline approach over the various classes of the heterogeneous MVTec benchmark dataset across all the metrics considered
A Secure Real-time Multimedia Streaming through Robust and Lightweight AES Encryption in UAV Networks for Operational Scenarios in Military Domain
Full text linkmultimodal data encryption and decryption for security applications in protected environments like espionage, situational awareness, monitoring, and counter-UAV. Data is captured from drones equipped with microphone arrays and cameras. This is performed by exploiting acoustic event analysis, video tracking, and recognition, performed on a ground station. All the communications are delivered in a secure data channel. Integrity and secrecy of the sensitive data acquired by drones must be guaranteed until the data is delivered in real-time from UAVs to the destination node. A possible data exploit may cause critical problems if the data is intercepted by malicious attackers. Being the drones equipped with low energy consuming devices with low computational power, like single-board-computers, a real-time lightweight application-level AES encryption, in addition to the MAC encryption of the wireless communication channel, has been considered. In the experiment, the encryption and decryption process has been optimized, even under adverse transmission conditions ensuring continuous data encryption even if some packets are lost or the connection is repeatedly dropped and reestablished
Combining Keypoint Clustering and Neural Background Subtraction for Real-time Moving Object Detection by PTZ Cameras
No full textDetection of moving objects is a topic of great interest in computer vision. This task represents a prerequisite
for more complex duties, such as classification and re-identification. One of the main challenges regards the
management of dynamic factors, with particular reference to bootstrapping and illumination change issues.
The recent widespread of PTZ cameras has made these issues even more complex in terms of performance due
to their composite movements (i.e., pan, tilt, and zoom). This paper proposes a combined keypoint clustering
and neural background subtraction method for real-time moving object detection in video sequences acquired
by PTZ cameras. Initially, the method performs a spatio-temporal tracking of the sets of moving keypoints to
recognize the foreground areas and to establish the background. Subsequently, it adopts a neural background
subtraction to accomplish a foreground detection, in these areas, able to manage bootstrapping and gradual
illumination changes. Experimental results on two well-known public datasets and comparisons with different
key works of the current state-of-the-art demonstrate the remarkable results of the proposed method
- …
