Search the repository

A+ | A- | | SLO | ENG

Abstract: Aims: With the ever-increasing survey speed of optical wide-field telescopes and the importance of discovering transients when they are still young, rapid and reliable source localization is paramount. We present AutoSourceID-Light (ASID-L), an innovative framework that uses computer vision techniques that can naturally deal with large amounts of data and rapidly localize sources in optical images. Methods: We show that the ASID-L algorithm based on U-shaped networks and enhanced with a Laplacian of Gaussian filter provides outstanding performance in the localization of sources. A U-Net network discerns the sources in the images from many different artifacts and passes the result to a Laplacian of Gaussian filter that then estimates the exact location. Results: Using ASID-L on the optical images of the MeerLICHT telescope demonstrates the great speed and localization power of the method. We compare the results with SExtractor and show that our method outperforms this more widely used method rapidly detects more sources not only in low and mid-density fields, but particularly in areas with more than 150 sources per square arcminute. The training set and code used in this paper are publicly available.
Keywords: astronomical databases, data analysis, image processing
Published in RUNG: 23.01.2023; Views: 596; Downloads: 0
This document has many files! More...

Keywords: Pitch processing, German, Italian, Deaccentuation, ERP, EEG, N400, Late Positivity, Given Information
Published in RUNG: 14.12.2021; Views: 1105; Downloads: 0
This document has many files! More...

Abstract: The use of non-literal language, like verbal irony, is deeply embedded in everyday communication and the ability to comprehend it changes across life. According to the echoic mention theory (Wilson & Sperber, 2012), understanding irony amounts to recognize a dissociative attitude. In the ‘Contextual echo’ example (Figure 1), Cynthia’s utterance “Tonight we gave a superb performance” is an example of irony. Cynthia is expressing a dissociative, mocking attitude towards the blatantly false proposition “Tonight we gave a superb performance”, that echoes the unfulfilled expectation that the concert would go well. The thought that is echoed can be “uttered”, like in the ‘Explicit echo’ example where the ironic utterance echoes the content explicitly expressed by Lea’s preceding statement; but the dissociative attitude can also target some implicitly communicated meaning, like in the ‘Implicated echo’ example, where the ironic utterance echoes the implicature of Lea’s statement, that is that they will sing well. Our first aim is to assess whether the echo’s degree of explicitness influence the processing of irony. Second, since research indicates that older adults sometimes struggle in understanding non-literal statements, like presupposition (Domaneschi & Di Paola 2019) or humor (Bischetti et al. 2019), we want to address the question of whether the processing of irony is more effortful in late adulthood and, if so, which underlying cognitive capacities might be responsible. Data collection is ongoing and the study is pre-registered on OSF (https://osf.io/94mys/?view_only=51fecb7acd694eca9b6b4d08cca02a26). Methods: The experiment requires the participation of 25 young adults (18-29-year-old) and 25 older adults (65-74-year-old). Participants will be administered a series of standardized tests to assess a) ToM (Faux Pas test) b) WM (Alpha span test) c) Autistic Quotient. The experimental study is a self-paced reading task. Each participant will be presented with stories adapted from the material of Spotorno & Noveck (2014): 15 ironic stories (5 with contextual echo, 5 with implicated echo and 5 with explicit echo), 5 literal stories, 10 decoys and 20 fillers (in a randomized order). Participants answer a yes/no comprehension question at the end of each story. An example of stories is given in Figure 1. Predictions: We expect overall slower reading time for ironic statements compared to literal ones and greater difficulties in the older adults group for ironic statements. We predict that our manipulation of the echo will have an effect on the processing of irony, and that reading times will be faster when the echo is explicit compared to when the echo is implicated (a stronger effect for older adults). We also expect that performance in our ToM task will predict reading times for ironic statements, with lower performance resulting in slower reading times. The presence of an implicated echo will exacerbate the difficulties. Moreover, we expect a positive correlation between the Autistic Quotient score and the difference between the reading times in the ironic and literal conditions. Finally, we expect that WM score will predict longer reading times for ironic statements when the implicitness of the echo poses higher cognitive demands. Analysis plan: First, we plan an evaluation of the group differences for neuropsychological data using a Wilcoxon signed- rank test. Then, we will proceed with a Pearson correlation coefficient test and analysis of variance to understand the relationship between the different measures (Clark, et al. 2010). The principal component analysis will be used to further assess their relationship. To understand the effect of the predictors on the reading time we will run a (Generalized) Linear Mixed-Effects Model with reading time as response variable, (Age Group x Type x Echo) as categorical predictors, test scores of neuropsychological data as continuous (or ordinal) predictors, and subject ID and items as random effects. All relevant interactions (both fixed and random) will also be assessed. The models will be fitted in R using the ‘lme4’ package (Bates et al. 2015). The (G)LMM will be simplified by removing one non-significant interaction at a time (and then, possibly non-significant main effects) on the basis of the Analysis of Deviance (LR Tests), until the optimal model is reached.
Keywords: irony, processing, aging
Published in RUNG: 22.09.2021; Views: 1201; Downloads: 77
Link to full text
This document has many files! More...

Abstract: Irony comprehension is a complex task that typically developing (TD) children reach around the age of six. Some scholars (Sullivan et al., 1995 a.o.) claimed that 2nd order Theory of Mind (ToM) skills are required to understand irony, but also linguistic abilities (which predict ToM development: Milligan, Astington & Dack, 2007) play a role (Filippova & Astington, 2008; Bosco & Gabbatore, 2017). Children with high-functioning autism (HFA) have intact linguistic abilities, but impaired social relations. Some children with ASD pass even 2nd order ToM tasks, even if they could be using compensatory verbalizing strategies (Fisher, Happé & Dunn, 2005; Happé, 1995; Tager-Flusberg, 2000). We tested irony comprehension in HFA children with the aim to assess the contribution of the factors that may facilitate it and disentangle their relationships. We analyzed responses to the key question to understand irony, i.e. questions on speaker’s meaning and attitude. HFA children showed a peculiar pattern: their accuracy on literal stories was at ceiling, demonstrating that they understood the task, but in irony comprehension they lag behind their TD peers matched for age and non-verbal IQ. Even if this result was not unexpected, given the impairment in social communication associated to HFA, our group of 26 HFA showed a somehow surprising bimodal distribution. Moreover, differently from the TD group, accuracy on ironic stories did not depend neither on age nor on (non-)verbal IQ in HFA participants. And conversely ToM skills played a significant role in irony understanding only for HFA children, and not for TD children.In order to account for these data, we can hypothesize that – in general – HFA children show an impairment in pragmatic inference abilities and in ToM abilities (see Loukusa & Moilanen, 2009 and Baron-Cohen, 2000 for a review), as found also in our sample of LP children. The HP children, on the other hand, might adopt a strategy, different from TD children, to respond correctly to ironic stories, as suggested by Wang et al. (2006). Interestingly, Pexman et al. (2011) found that HFA children, who did not differ in accuracy with respect to TD controls, applied a different processing strategy for irony comprehension, and hypothesized that they resort to a more rule-based strategy, with an intellectual-style approach to compensate their social deficits. Our bimodal distribution could then identify two classes of HFA participants: the LP ones correspond to the pragmatic-impaired profile of HFA, the HP ones on the other hand could be using a compensatory strategy, since they could have been trained to recognize irony and/or other persons’ states of mind (ToM) during speech therapy sessions, as reported in Persicke et al. (2013). Future research should investigate this hypothesis more in depth, with a longitudinal study of HFA participants after a training in irony comprehension, in ToM metarepresentational abilities, and without training.
Keywords: irony processing, autism developmental disorder, theory of mind, IQ, experimental pragamtics
Published in RUNG: 21.09.2021; Views: 1381; Downloads: 52
Link to full text
This document has many files! More...

Abstract: Technological advancements in the field of multimedia have made access to information effortless. As information is available easily to everyone so it is quite essential to shield secretive data. This paper focuses on encrypting videos using confusion and diffusion at various levels. Chaotic skew tent map and nested piecewise linear chaotic map which are categorized under piecewise linear maps have been used for the processing. Two skew tent maps are cross coupled with each other to generate random values. The binary sequences are generated by comparing the outputs of the cross coupled piecewise linear chaotic maps. The generated binary sequence is used for bit level diffusion. Further, pixels are shuffled to provide higher level of encryption. The ciphered video is of the same size as that of plain video. The proposed scheme is of low complexity and provides high level encryption. The security analysis of the technique has been carried out and better results have been found regarding the statistical and differential attacks. The code development has been done in MATLAB and the scheme is applicable in areas like Video Encryption, Steganography and Digital Watermarking on various lightweight devices.
Keywords: video processing, multimedia encryption, chaotic maps, lightweight device security
Published in RUNG: 01.04.2021; Views: 1515; Downloads: 0
This document has many files! More...