Fake audio detection using Hierarchical Representations Learning and Spectrogram Features

Author(s):  
Mohsan Ali ◽  
Alina Sabir ◽  
Mehdi Hassan
2021 ◽  
Author(s):  
Jeffrey Zheng ◽  
Minghan Zhu ◽  
Mu Qiao ◽  
Yang Zhou

Abstract In this paper, comprehensive cases are visualized by {C1, C2, C3, C4} four modules of the MAS. Four sets of SARS-CoV-2 genomes with 20, 128, 381 and 1337 cases were processed to be represented as a series of genomic index maps in four entropies: combinatorial, integrated, mean and topological entropies, respectively. Typical samples and explanations are described. Multiple levels of hierarchical representations are illustrated.


Retos ◽  
2021 ◽  
pp. 299-309
Author(s):  
Lorena Velasco-Santos ◽  
José Luis Pastor Pradillo ◽  
David Blanco-Alcántara ◽  
Alfredo Jiménez Eguizábal

  El presente artículo analiza los valores del cuerpo de 536 estudiantes de 1º Bachillerato en Castilla y León en el curso 2017/18 en función de sus variables de perfil sexo, modalidad de bachillerato, naturaleza, titularidad y provincia de pertenencia del centro de estudios. El objetivo es describir tendencias generales asociadas a dichas variables de perfil. El instrumento de medida es el test elaborado por Casares y Collados (1998) que clasifica en 10 los valores del cuerpo en base a 250 ítems agrupados en bloques de 25. El tratamiento de datos se lleva a cabo mediante un análisis de valores medios. Los resultados obtenidos son mostrados en forma de tablas de puntuaciones, gráficos y representaciones jerárquicas. Se calculan estadísticos descriptivos y estadística inferencial. Se observan diferencias estadísticamente significativas asociadas a la variable de perfil sexo y, secundariamente, provincia de pertenencia del centro de estudios. Asimismo, destaca el valor placer como el más agradable en todas las clasificaciones y el valor religioso el menos, exceptuando lo ocurrido en Soria donde se torna incluso positivo. Todo ello describe la realidad en que los estudiantes de 1º bachillerato conciben su cuerpo y le conceden su valor, lo que abre nuevos flancos críticos en la educación a través de los valores del cuerpo.  Abstract. This article analyzes the body values of 536 First Year Bachillerato students in the Castilla y León region of Spain during the academic year 2017/18 according to their special characteristics such as gender, Bachillerato option, personality, qualifications and province administering the educational establishment they attend. The aim is to describe general trends associated with these profile variables. The instrument of measurement is the test devised by Casares and Collados (1998), which divides body values into 10 categories. It is based on 250 items classified into 10 groups of 25 defining the body values. Data processing is carried out through an analysis of average values. The results obtained are shown in the form of score tables, graphs and hierarchical representations. Descriptive statistics and inferential statistics are calculated. Significant differences associated to the variable of gender profile and, secondarily, province of belonging to the study center are highlighted. Likewise, the pleasure value stands out as the most pleasant in all classifications and the religious value the least, except for what happened in Soria where it becomes even positive. The procedure of preferred values ratifies the pleasure value as the most pleasant, however the religious value is rarely the least preferred. All this describes the reality in which the students of the 1st baccalaureate conceive their body and give it their value, which opens new critical flanks in education through the values ​​of the body.


Author(s):  
Lianli Gao ◽  
Zhilong Zhou ◽  
Heng Tao Shen ◽  
Jingkuan Song

Image edge detection is considered as a cornerstone task in computer vision. Due to the nature of hierarchical representations learned in CNN, it is intuitive to design side networks utilizing the richer convolutional features to improve the edge detection. However, there is no consensus way to integrate the hierarchical information. In this paper, we propose an effective and end-to-end framework, named Bidirectional Additive Net (BAN), for image edge detection. In the proposed framework, we focus on two main problems: 1) how to design a universal network for incorporating hierarchical information sufficiently; and 2) how to achieve effective information flow between different stages and gradually improve the edge map stage by stage. To tackle these problems, we design a consecutive bottom-up and top-down architecture, where a bottom-up branch can gradually remove detailed or sharp boundaries to enable accurate edge detection and a top-down branch offers a chance of error-correcting by revisiting the low-level features that contain rich textual and spatial information. And attended additive module (AAM) is designed to cumulatively refine edges by selecting pivotal features in each stage. Experimental results show that our proposed methods can improve the edge detection performance to new records and achieve state-of-the-art results on two public benchmarks: BSDS500 and NYUDv2.


2017 ◽  
Vol 17 (3) ◽  
pp. 13 ◽  
Author(s):  
Odelia Schwartz ◽  
Luis Gonzalo Sanchez Giraldo

2019 ◽  
Vol 8 (1) ◽  
pp. 46 ◽  
Author(s):  
François Merciol ◽  
Loïc Faucqueur ◽  
Bharath Damodaran ◽  
Pierre-Yves Rémy ◽  
Baudouin Desclée ◽  
...  

Land cover mapping has benefited a lot from the introduction of the Geographic Object-Based Image Analysis (GEOBIA) paradigm, that allowed to move from a pixelwise analysis to a processing of elements with richer semantic content, namely objects or regions. However, this paradigm requires to define an appropriate scale, that can be challenging in a large-area study where a wide range of landscapes can be observed. We propose here to conduct the multiscale analysis based on hierarchical representations, from which features known as differential attribute profiles are derived over each single pixel. Efficient and scalable algorithms for construction and analysis of such representations, together with an optimized usage of the random forest classifier, provide us with a semi-supervised framework in which a user can drive mapping of elements such as Small Woody Features at a very large area. Indeed, the proposed open-source methodology has been successfully used to derive a part of the High Resolution Layers (HRL) product of the Copernicus Land Monitoring service, thus showing how the GEOBIA framework can be used in a big data scenario made of more than 38,000 Very High Resolution (VHR) satellite images representing more than 120 TB of data.


Sign in / Sign up

Export Citation Format

Share Document