scholarly journals Accurate Multilevel Classification for Wildlife Images

2021 ◽  
Vol 2021 ◽  
pp. 1-11
Author(s):  
Francisco Gomez-Donoso ◽  
Félix Escalona ◽  
Ferran Pérez-Esteve ◽  
Miguel Cazorla
Keyword(s):  
Data Augmentation ◽  
General Concept ◽  
Specific Class ◽  
Human Knowledge ◽  
Specific Element ◽  
Cascade Classifiers ◽  
Input Size ◽  
Exhaustive Study ◽  
Taxonomic Tree ◽  
Multilevel Classification

The most common approaches for classification rely on the inference of a specific class. However, every category could be naturally organized within a taxonomic tree, from the most general concept to the specific element, and that is how human knowledge works. This representation avoids the necessity of learning roughly the same features for a range of very similar categories, and it is easier to understand and work with and provides a classification for each abstraction level. In this paper, we carry out an exhaustive study of different methods to perform multilevel classification applied to the task of classifying wild animals and plant species. Different convolutional backbones, data setups, and ensembling techniques are explored to find the model which provides the best performance. As our experimentation remarks, in order to achieve the best performance on the datasets that are arranged in a tree-like structure, the classifier must feature an EfficientNetB5 backbone with an input size of 300 × 300 px, followed by a multilevel classifier. In addition, a Multiscale Crop data augmentation process must be carried out. Finally, the accuracy of this setup is a 62% top-1 accuracy and 88% top-5 accuracy. The architecture could benefit for an accuracy boost if it is involved in an ensemble of cascade classifiers, but the computational demand is unbearable for any real application.

scholarly journals Generative Adversarial Networks to Improve the Robustness of Visual Defect Segmentation by Semantic Networks in Manufacturing Components

2021 ◽  
Vol 11 (14) ◽  
pp. 6368
Author(s):  
Fátima A. Saiz ◽  
Garazi Alfaro ◽  
Iñigo Barandiaran ◽  
Manuel Graña
Keyword(s):  
Ad Hoc ◽  
Data Augmentation ◽  
Semantic Network ◽  
Semantic Networks ◽  
Stereo Image ◽  
Generative Adversarial Networks ◽  
Specific Class ◽  
Adversarial Networks ◽  
Augmentation Techniques ◽  
Image Acquisition System

This paper describes the application of Semantic Networks for the detection of defects in images of metallic manufactured components in a situation where the number of available samples of defects is small, which is rather common in real practical environments. In order to overcome this shortage of data, the common approach is to use conventional data augmentation techniques. We resort to Generative Adversarial Networks (GANs) that have shown the capability to generate highly convincing samples of a specific class as a result of a game between a discriminator and a generator module. Here, we apply the GANs to generate samples of images of metallic manufactured components with specific defects, in order to improve training of Semantic Networks (specifically DeepLabV3+ and Pyramid Attention Network (PAN) networks) carrying out the defect detection and segmentation. Our process carries out the generation of defect images using the StyleGAN2 with the DiffAugment method, followed by a conventional data augmentation over the entire enriched dataset, achieving a large balanced dataset that allows robust training of the Semantic Network. We demonstrate the approach on a private dataset generated for an industrial client, where images are captured by an ad-hoc photometric-stereo image acquisition system, and a public dataset, the Northeastern University surface defect database (NEU). The proposed approach achieves an improvement of 7% and 6% in an intersection over union (IoU) measure of detection performance on each dataset over the conventional data augmentation.

scholarly journals La distribución de competencias entre el Estado y las Comunidades Autónomas en materia de agricultura

2010 ◽  
Author(s):  
Manuel Rodríguez Portugués
Keyword(s):  
General Concept ◽  
Power Sharing ◽  
Constitutional Court ◽  
The European Union ◽  
Legal Context ◽  
El Sistema ◽  
The Common ◽  
Autonomous Regions ◽  
Exhaustive Study ◽  
Constitutional Jurisprudence

<p align="justify">En este trabajo se estudia el sistema de distribución de competencias entre el Estado y las Comunidades Autónomas sobre agricultura. Con él se pretende ofrecer un estudio exhaustivo de esta vasta materia teniendo en cuenta y sistematizando la abundante jurisprudencia constitucional que se ha ido formando desde el inicio del Estado de las Autonomías hasta la actualidad, así como el nuevo contexto jurídico abierto por la reciente Sentencia del Tribunal Constitucional 31/2010, de 28 de junio, sobre el Estatuto de Autonomía de Cataluña, o por la ya consolidada posición de la Comunidad Autónoma de Canarias como región ultraperiférica de la Unión Europea en relación a la aplicación del llamado “primer pilar” de la Política Agrícola Común en su territorio. Además, y sobre la base de todo ello, se formula una propuesta de concepto general de “agricultura” que sea válido para deslindar esta materia de otras afines a efectos competenciales.</p> <p align="justify"><b>In this study the power-sharing system for agriculture between the State and Spanish autonomous regions will be examined. The study aims to offer an exhaustive study of this vast subject, taking into account and systemising the abundant constitutional jurisprudence that has been forming since the creation of the autonomous regions until the present day, as well as the new legal context opened by the recent ruling of the Spanish Constitutional Court (31/2010, 28th June) over the Catalan Statute of Autonomy, and for the already consolidated position of the autonomous region of the Canary Islands as the outermost region of the European Union in relation to the application of what is known as the ‘first pier’ of the Common Agricultural Policy within its territory. Also, and based on the above, a proposal regarding the general concept of ‘agriculture’ will be formulated, one which is valid to separate this subject from other matters relative to the effects of competencies.</p>

scholarly journals Detection of Invasive Species in Wetlands: Practical DL with Heavily Imbalanced Data

2020 ◽  
Vol 12 (20) ◽  
pp. 3431 ◽  
Author(s):  
Mariano Cabezas ◽  
Sarah Kentsch ◽  
Luca Tomhave ◽  
Jens Gross ◽  
Maximo Larry Lopez Caceres ◽  
...  
Keyword(s):  
Data Augmentation ◽  
Imbalanced Data ◽  
Practical Interest ◽  
True Positive Rate ◽  
Ease Of Use ◽  
Aerial Images ◽  
Specific Class ◽  
Final Layer ◽  
Random Weights ◽  
Positive Rate

Deep Learning (DL) has become popular due to its ease of use and accuracy, with Transfer Learning (TL) effectively reducing the number of images needed to solve environmental problems. However, this approach has some limitations which we set out to explore: Our goal is to detect the presence of an invasive blueberry species in aerial images of wetlands. This is a key problem in ecosystem protection which is also challenging in terms of DL due to the severe imbalance present in the data. Results for the ResNet50 network show a high classification accuracy while largely ignoring the blueberry class, rendering these results of limited practical interest to detect that specific class. Moreover, by using loss function weighting and data augmentation results more akin to our practical application, our goals can be obtained. Our experiments regarding TL show that ImageNet weights do not produce satisfactory results when only the final layer of the network is trained. Furthermore, only minor gains are obtained compared with random weights when the whole network is retrained. Finally, in a study of state-of-the-art DL architectures best results were obtained by the ResNeXt architecture with 93.75 True Positive Rate and 98.11 accuracy for the Blueberry class with ResNet50, Densenet, and wideResNet obtaining close results.

Love as Origin of Science

2021 ◽  
Vol 63 (4) ◽  
pp. 455-472
Author(s):  
Heung Myung Oh
Keyword(s):  
Philosophy Of Science ◽  
Scientific Knowledge ◽  
General Concept ◽  
Human Knowledge ◽  
Theory Of Science ◽  
Alternative Approach ◽  
Rigorous Model ◽  
Externalist Approach ◽  
Objective Validity ◽  
Ultimate Ground

Summary The approaches to the possibility of theology as science are divided roughly into three types: first, the internalist approach which rejects any attempt to verify the objective validity of revelation under the general concept of science. Second, the externalist approach which demands the verification of objective validity of revelatory truth. Third, the inclusivist approach which seeks the scientificity of theology from a hermeneutic perspective. Outlining the crucial points and limits of these approaches and replacing the question about theology as science with a theological reexamination of the possibility of science in general, this paper tries to suggest an alternative approach by establishing the possibility of scientific knowledge in general from the trinitarian perspective. Under this reformulation of the question, the philosophy of science set forth by Fichte as the most rigorous model of theory of science is critically explored. In conclusion, it is argued that the ultimate ground of all human knowledge and science consists in the eternal divine love and trust in it.

scholarly journals Detecting Incongruity between News Headline and Body Text via a Deep Hierarchical Encoder

2019 ◽  
Vol 33 ◽  
pp. 791-800
Author(s):  
Seunghyun Yoon ◽  
Kunwoo Park ◽  
Joongbo Shin ◽  
Hongjun Lim ◽  
Seungpil Won ◽  
...  
Keyword(s):  
Body Text ◽  
Data Augmentation ◽  
The Body ◽  
False Information ◽  
News Stories ◽  
Text Input ◽  
Input Size ◽  
Hierarchical Architectures ◽  
Textual Representation ◽  
News Headlines

Some news headlines mislead readers with overrated or false information, and identifying them in advance will better assist readers in choosing proper news stories to consume. This research introduces million-scale pairs of news headline and body text dataset with incongruity label, which can uniquely be utilized for detecting news stories with misleading headlines. On this dataset, we develop two neural networks with hierarchical architectures that model a complex textual representation of news articles and measure the incongruity between the headline and the body text. We also present a data augmentation method that dramatically reduces the text input size a model handles by independently investigating each paragraph of news stories, which further boosts the performance. Our experiments and qualitative evaluations demonstrate that the proposed methods outperform existing approaches and efficiently detect news stories with misleading headlines in the real world.

Mind wandering as data augmentation: How mental travel supports abstraction

2020 ◽  
Vol 43 ◽  
Author(s):  
Myrthe Faber
Keyword(s):  
Machine Learning ◽  
Data Augmentation ◽  
Mental Content ◽  
Mind Wandering ◽  
Theoretical Framework ◽  
Important Addition

Abstract Gilead et al. state that abstraction supports mental travel, and that mental travel critically relies on abstraction. I propose an important addition to this theoretical framework, namely that mental travel might also support abstraction. Specifically, I argue that spontaneous mental travel (mind wandering), much like data augmentation in machine learning, provides variability in mental content and context necessary for abstraction.

Fusion and Fission Processes in Exocytosis: Possible Roles for Synexin and Osmotic Lysis in the Two Events

1980 ◽  
Vol 38 ◽  
pp. 594-597
Author(s):  
H.B. Pollard ◽  
C.E. Creutz ◽  
C.J. Pazoles ◽  
J.H. Scott
Keyword(s):  
Small Molecules ◽  
Chromaffin Cells ◽  
Adrenal Glands ◽  
General Concept ◽  
Extracellular Calcium ◽  
Large Protein ◽  
Granule Membrane ◽  
Osmotic Lysis ◽  
Per Se ◽  
Chemical Evidence

Exocytosis is a general concept describing secretion of enzymes, hormones and transmitters that are otherwise sequestered in intracellular granules. Chemical evidence for this concept was first gathered from studies on chromaffin cells in perfused adrenal glands, in which it was found that granule contents, including both large protein and small molecules such as adrenaline and ATP, were released together while the granule membrane was retained in the cell. A number of exhaustive reviews of this early work have been published and are summarized in Reference 1. The critical experiments demonstrating the importance of extracellular calcium for exocytosis per se were also first performed in this system (2,3), further indicating the substantial service given by chromaffin cells to those interested in secretory phenomena over the years.

Deep neural networks trained with heavier data augmentation learn features closer to representations in hIT

2018 ◽  
Author(s):  
Alex Hernández-García ◽  
Johannes Mehrer ◽  
Nikolaus Kriegeskorte ◽  
Peter König ◽  
Tim C. Kietzmann
Keyword(s):  
Neural Networks ◽  
Deep Neural Networks ◽  
Data Augmentation

Ale na razie tu cisza” – fotografie sprzed katastrofy w wybranych zapisach literackich (Stefan Chwin, Ryszard Kapuściński, Wisława Szymborska)

2020 ◽  
pp. 65-80
Author(s):  
Magdalena Strąk
Keyword(s):  
Time Perspective ◽  
Terrorist Attack ◽  
First World War ◽  
General Concept ◽  
Literary Texts ◽  
World War ◽  
The Past ◽  
The First World War ◽  
History Of ◽  
First World

The work aims to show a peculiar perspective of looking at photographs taken on the eve of the broadly understood disaster, which is specified in a slightly different way in each of the literary texts (Stefan Chwin’s autobiographical novel Krótka historia pewnego żartu [The brief history of a certain joke], a poem by Ryszard Kapuściński Na wystawie „Fotografia chłopów polskich do 1944 r.” [At an exhibition “The Polish peasants in photographs to 1944”] and Wisława Szymborska’s Fotografia z 11 września [Photograph from September 11]) – as death in a concentration camp, a general concept of the First World War or a terrorist attack. Upcoming tragic events – of which the photographed people are not yet aware – become for the subsequent recipient an inseparable element of reality contained in the frame. For the later observers, privileged with time perspective, the characters captured in the photograph are already victims of the catastrophe, which in reality was not yet recorded by the camera. It is a work about coexistence of the past and future in the field of photography.

Sign in / Sign up

Export Citation Format

Share Document