Search results for "back"

A system based on neural architectures for the reconstruction of 3-D shapes from images

1991

The connectionist approach to the recovery of 3-D shape information from 2-D images developed by the authors, is based on a system made up by two cascaded neural networks. The first network is an implementation of the BCS, an architecture which derives from a biological model of the low level visual processes developed by Grossberg and Mingolla: this architecture extracts a sort of brightness gradient map from the image. The second network is a backpropagation architecture that supplies an estimate of the geometric parameters of the objects in the scene under consideration, starting from the outputs of the BCS. A detailed description of the system and the experimental results obtained by si…

A greedy perturbation approach to accelerating consensus algorithms and reducing its power consumption

2011

The average consensus is part of a family of algorithms that are able to compute global statistics by only using local data. This capability makes these algorithms interesting for applications in which these distributed philosophy is necessary. However, its iterative nature usually leads to a large power consumption due to the repetitive communications among the iterations. This drawback highlights the necessity of minimizing the power consumption until consensus is reached. In this work, we propose a greedy approach to perturbing the connectivity graph, in order to improve the convergence time of the consensus algorithm while keeping bounded the power consumption per iteration step. These …

No evidence of systematic pre-emptive loggings after notifying landowners of their lands’ conservation potential

2020

Landowners can intentionally impair biodiversity values occurring on their land to pre-empt biodiversity protection. This often leads to significant negative effects on biodiversity. We studied whether landowners in Finland engaged in pre-emptive loggings after they were notified that their wooded mires are candidate sites for a mire protection program. After the notification, harvesting rates of the candidate wooded mires were significantly lower compared to harvesting rates of similar but non-candidate wooded mires. Annual and monthly harvesting rates indicated that notifying landowners of the conservation potential did not launch systematic pre-emptive logging behavior. Nevertheless, par…

2021

Objectives: Energy drink consumption among adolescents has become a notable global phenomenon, and has been associated with numerous negative health outcomes. In order to understand the popularity of energy drinks among adolescents, and to target interventions, it is important to identify the determinants underpinning consumption.Methods: The nationally representative data (cross-sectional) were drawn from the Health Behaviour in School-aged Children (HBSC) surveys, conducted in 2014 and 2018, each comprising 13- and 15-year-old Finnish adolescents (n = 7405).Results: Weekly energy drink consumption increased among Finnish adolescents between 2014 (18.2%) and 2018 (24.4%), especially among …

Ecological restoration in contaminated soils of Kokdzhon phosphate mining area (Zhambyl region, Kazakhstan)

2016

Abstract This study provides results of an ecological restoration activity performed in the Kokdzhon phosphate mining of the Zhambyl region (a semi-desert mining area of Kazakhstan). The test area was made by a quarry-hole, about 1.25 hectares wide, that was preliminarily filled with the earthy material of a dump and, subsequently, levelled. The total volume of human transported material (HTM) used was 700 tonnes. Soil samples were taken from 0 to 30 cm depth at 23 sampling sites and several soil quality parameters were determined. In the test area 750 tree species were planted in suitable dug holes: (150 seedlings of Russian Olive; 150 seedlings of Black Saxaul); 150 seedlings of Androsov …

Towards a Non-Intrusive Context-Aware Speech Quality Model

2020

Understanding how humans judge perceived speech quality while interacting through Voice over Internet Protocol (VoIP) applications in real-time is essential to build a robust and accurate speech quality prediction model. Speech quality is degraded in the presence of background noise reducing the Quality of Experience (QoE). Speech Enhancement (SE) algorithms can improve speech quality in noisy environments. The publicly available NOIZEUS speech corpus contains speech in environmental background noise babble, car, street, and train at two Signal-to-noise ratio (SNRs) 5dB and 10dB. Objective Speech Quality Metrics (OSQM) are used to monitor and measure speech quality for VoIP applications. Th…

Ethnicity and Educational Inequality: An investigation of School Experience in Australia and France. Ethnicité et inégalité scolaire : une enquête su…

2008

This thesis examines the contribution of ‘ethnic' background to the school experiences of educationally and socially disadvantaged students in the senior years of high school (n=927). To investigate the role both of ethnic identification and its interplay with institutional factors, a comparative analysis of secondary student experiences in two national settings was undertaken. The case of Turkish-background students in Australia and France suggests that the influences of ethnic identity are thoroughly transformed from one setting to the other by distinctive pedagogical structures. Streaming and severe academic judgement in France lower academic self-esteem, while creating resentment and so…

Deep CNN-ELM Hybrid Models for Fire Detection in Images

2018

In this paper, we propose a hybrid model consisting of a Deep Convolutional feature extractor followed by a fast and accurate classifier, the Extreme Learning Machine, for the purpose of fire detection in images. The reason behind using such a model is that Deep CNNs used for image classification take a very long time to train. Even with pre-trained models, the fully connected layers need to be trained with backpropagation, which can be very slow. In contrast, we propose to employ the Extreme Learning Machine (ELM) as the final classifier trained on pre-trained Deep CNN feature extractor. We apply this hybrid model on the problem of fire detection in images. We use state of the art Deep CNN…

Structural highs formation and their relationship to sedimentary basins in the north Sicily continental margin (southern Tyrrhenian Sea): implication…

2005

Two high-penetration seismic profiles across the continental margin of north Sicily (southern Tyrrhenian Sea) have been interpreted to construct a crustal section across the margin and to provide a three-dimensional perspective on the upper crustal setting of the Solunto High and its structural relationships with the adjacent Cefalù Basin. Crystalline rocks of the Kabilian-Calabrian Units (KCU) are recognized in the Solunto High region and farther to the north. The KCU body is tectonically superimposed onto sedimentary rocks pertaining to the Sicilian-Maghrebian Units (SMU) and the Solunto High corresponds to the thrust sheet culmination. Its occurrence in the Solunto High demonstrates that…

On numerical simulation of the continuous casting process

1988

In this paper a steady-state nonlinear parabolic-type model, which simulates the multiphase heat transfer during solidification in continuous casting, is presented. An enthalpy formulation is used and we apply a FE-method in space and an implicit Euler method in time. A detailed solution algorithm is presented. We compute the temperature distributions in the strand when the boundary conditions (mold/spray cooling) on the strand surface are known. The numerical model gives thereby a good basis for the testing of new designs of continuous-casting machines. An application of the model to continuous casting of billets is presented.