1,721,037 research outputs found
RANSAC-LEL: An optimized version with Least Entropy Like Estimators
No full textThe paper proposes a robust estimation method which implements, in cascade, two algorithms: (i) a Random Sample and Consensus (RANSAC) algorithm and (ii) a novel nonlinear prediction error estimator minimizing a cost function inspired by the mathematical definition of Gibbs entropy. The minimization of the nonlinear cost function allows to refine the Consensus Set found with standard RANSAC in order to reach optimal estimates of geometric transformation parameters under image stitching context. The method has been experimentally tested and compared with a standard RANSAC-MSAC algorithm where noticeable improvements are recorded in terms of computational complexity and quality of the stitching process, namely of the mean squared symmetric re-projection error
Robust 3D Plane Estimation for Autonomous Vehicle Applications
No full textA novel plane estimation algorithm from 3D range data is presented. The proposed solution is based on the minimization of a nonlinear prediction error cost function inspired by the mathematical definition of Gibbs' entropy. The method has been experimentally tested and compared with a standard implementation of the RANSAC algorithm. Results suggest that the proposed approach has the potential of performing better in terms of precision and reliability while requiring a lower computational effort
Application of a gas sensors array to the detection of fuel as contamination defect in engine oil
No full textAbstract — In this work we proposed a system based on metal
oxide gas micro-sensors to estimate diesel or gasoline contamination in different engine oil samples. The gas-sensing
layers (undoped, Pt, Pd, Rh-doped SnO2 , In2 O3 and mixed
In2 O3 -SnO2 ) have been synthetized by the sol-gel method and
deposited by spin-coating onto 2mm x 2mm silicon substrates
equipped by Pt heater on the back and Pt interdigitated
electrodes on the front. The sensor array has been exposed to
no-used and used commercial engine oil samples contaminated
with different amounts of unburned fuel. The results of data
analysis (DWT-based feature extraction, PCA and Gaussian
mixture model classifier (GMM)) showed that different fuel
contaminated used engine oils can be discriminated and
successfully classified by the sensor array
Integrating microwave reflectometry and deep learning imaging for in-vivo skin cancer diagnostics
No full textSkin cancer is a major global health concern, with rising incidence rates. However, current diagnostic methods often lack objectivity, speed, and non-invasiveness. To overcome this limitation, this work proposes a novel approach combining microwave reflectometry (MR) with artificial intelligence (AI) techniques for the diagnosis of in-vivo skin cancer. In particular, MR exploits the dielectric properties of biological tissues, revealing chemical-physical differences in normal skin and benign/malignant lesions. To better improve the diagnostic performance, MR analysis is integrated with AI algorithms, particularly those based on deep learning (DL) and convolutional neural networks (CNNs), which analyze dermoscopic skin images and identify asymmetry, irregular borders, abnormal colorations, and other skin cancer indicators. Combining MR with AI image analysis provides a comprehensive diagnostic approach, as MR informs on tissue dielectric composition, while AI analyzes lesion-specific details, offering more accurate and timely assessments. This combined approach promises early skin cancer detection and could significantly impact clinical practice
A UAV-Based Visual Tracking Algorithm for Sensible Areas Surveillance
No full textUnmanned aerial vehicles (UAVs) are an active research field since several years. They can be applied in a large variety of different scenarios, and supply a test bed to investigate several unsolved problems such as path planning, control and navigation. Furthermore, with the availability of low cost, robust and small video cameras, UAV video has been one of the fastest growing data sources in the last couple of years. In other words, object detection and tracking as well as visual navigation has recently received a lot of attention. This paper proposes an advanced technology framework that, through the use of UAVs, allows to supervise a specific sensible area (i.e. traffic monitoring, dangerous zone and so on). In particular, one of the most cited real-rime visual tracker proposed in the literature, Struck, is applied on video sequences tipically supplied by UAVs equipped with amonocular camera. Furthermore in this paper is investigated on the feasibility to graft different features characterization into the original tracking architecture (replacing the orginal ones). The used feature extraction methods are based on Local Binary Pattern (LBP) and Histogram of Oriented Gradients (HOG). Objects to be tracked could be selected manually or by means of advanced detection technique based, for example, on change detection or template matching strategies. The experimental results on well known benchmark sequences show as these features replacing improve the overall performances of the original considered real-time visual tracker
A Microservices Architecture based on a Deep-learning Approach for an Innovative Fruition of Art and Cultural Heritage
Full text linkTechnological innovations have resulted in a digital transformation in a variety of fields, including culture and tourism. We propose an innovative and personalized solution to benefit art and cultural heritage in indoor and outdoor environments by combining Internet of Things-enabled technologies and deep learning-based approaches. A recent Convolutional Neural Network (CNN) architecture to jointly perform local feature detection and description has been adapted and exploited for the first time for image matching in the cultural heritage application context. The performance validation of the proposed system shows that the proposed modular architecture ensures a very low error rate and excellent response time up to 2000 user visits in 700 seconds. The validation of the computer vision module shows as the proposed CNN based feature extraction approach improves image matching performance, especially in poorly textured object areas reaching a F1-Score of 0.9907 (against the 0.9679 obtained by traditional gradient based approaches) on the challenging dataset of images taken from 4 different historical sites and a F1-Score of 0.9807 (against the 0.9798 obtained by traditional approaches) on a public benchmark dataset of artworks
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Optimized Connected Components Labeling with Pixel Prediction
No full textIn this paper we propose a new paradigm for connected components labeling, which employs a general approach to minimize the number of memory accesses, by exploiting the information provided by already seen pixels, removing the need to check them again. The scan phase of our proposed algorithm is ruled by a forest of decision trees connected into a single graph. Every tree derives from a reduction of the complete optimal decision tree. Experimental results demonstrated that on low density images our method is slightly faster than the fastest conventional labeling algorithms
Variations on the Author
Full text link“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
- …
