scholarly journals Deep learning solution for children long-term identification

2020 ◽  
Author(s):  
Nikolajs Bumanis ◽  
◽  
Gatis Vitols ◽  
Irina Arhipova ◽  
Inga Meirane ◽  
...  
Keyword(s):  
Deep Learning ◽  
Object Tracking ◽  
Business Processes ◽  
Feature Analysis ◽  
Educational Institutions ◽  
Popular Method ◽  
Automated Learning ◽  
Identification Analysis ◽  
Term Identification

Deep learning algorithms are becoming default solution for application in business processes where recognition, identification and automated learning are involved. For human identification, analysis of various features can be applied. Face feature analysis is most popular method for identification of person in various stages of life, including children and infants. The aim of this research was to propose deep learning solution for long-term identification of children in educational institutions. Previously proposed conceptual model for long-term re-identification was enhanced. The enhancements include processing of unexpected persons’ scenarios, knowledge base improvements based on results of supervised and unsupervised learning, implementation of video surveillance zones within educational institutions and object tracking results’ data chaining between multiple logical processes. Object tracking results are the solution we found for long-term identification realization.

Current situation with the core collection of spring triticale of the National Center of Plant Genetic Resources of Ukraine

2018 ◽  
pp. 75-85
Author(s):  
S.V. Chernobai ◽  
V.K. Riabchun ◽  
T.B. Kapustina ◽  
V.S. Melnyk ◽  
O.E. Shchechenko
Keyword(s):  
Genetic Resources ◽  
Plant Genetic Resources ◽  
Plant Production ◽  
Educational Institutions ◽  
Breeding Lines ◽  
Growing Season Length ◽  
Long Term Storage ◽  
Foreign Countries ◽  
Spring Triticale

Goal. To build up a spring triticale genetic bank to provide breeding, scientific and educational institutions with initial material and to preserve the existing diversity. To update the database of accessions with a set of valuable economic and morphological features. Results and discussion. The methodology and results of the collection formation and evaluation of spring triticale accessions in the National Center for Plant Genetic Resources of Ukraine of Plant Production Institute nd. a V. Ya. Yuriev are presented. The formed collection includes 1,935 accessions from 27 countries: 42 varieties and 1,478 breeding lines from Ukraine, 92 varieties and 248 lines from foreign countries and also 75 genetic lines. The collection was formed by major valuable economic features (plant height, growing season length, spike threshing, yield, 1000-grain weight, disease resistance, technological properties, etc.). Accessions with the majority of morpho-biological and valuable economic features were selected. All the accessions in the collection are certificated. 1,762 accessions were packed for storage into the National Depository; 1,507 of them were packed for long-term storage. Conclusions. The gene pool of spring triticale from the collection of the Gene Bank of Plants of Ukraine is widely used for breeding. This allows conducting hybridization of genetically and ecologically remote forms with various expressions of features and obtaining whole new breeding material. Involvement of collection accessions in breeding allows generating new genetic sources of valuable economic features.

Marketing For Sustainability: A Conceptual Framework

2018 ◽  
Vol 8 (1) ◽  
Author(s):  
Dr.Pankaj Jain
Keyword(s):  
Social Marketing ◽  
Business Processes ◽  
Green Marketing ◽  
Marketing Social ◽  
Marketing Systems ◽  
Sustainable Marketing ◽  
Marketing Activities ◽  
The Mind ◽  
Marketing Principles

This paper is an attempt to put forward a roadmap to attain sustainable marketing through social marketing, green marketing and critical marketing. Social Marketing is an approach to decide the marketing strategies and activities keeping society’s long term welfare in the mind. Social and ethical concerns are at the centre of social marketing. Green Marketing is an approach to develop and market environmentally safer products and services in and introducing sustainability efforts in various marketing and business processes. At last, Critical Marketing is an approach that calls for analyzing marketing principles, techniques and theory using a critical theory based approach. This approach helps in regulating and controlling marketing activities with a focus on sustainability as it challenges and questions the existing capitalist and marketing systems so as to achieve a more sustainable marketing system.

scholarly journals Domain randomization-enhanced deep learning models for bird detection

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Xin Mao ◽  
Jun Kang Chow ◽  
Pin Siang Tan ◽  
Kuan-fu Liu ◽  
Jimmy Wu ◽  
...  
Keyword(s):  
Deep Learning ◽  
Continuous Monitoring ◽  
Bird Species ◽  
Training Data ◽  
Learning Models ◽  
Fine Grained ◽  
Bird Detection ◽  
Relationship Of ◽  
The Relationship

AbstractAutomatic bird detection in ornithological analyses is limited by the accuracy of existing models, due to the lack of training data and the difficulties in extracting the fine-grained features required to distinguish bird species. Here we apply the domain randomization strategy to enhance the accuracy of the deep learning models in bird detection. Trained with virtual birds of sufficient variations in different environments, the model tends to focus on the fine-grained features of birds and achieves higher accuracies. Based on the 100 terabytes of 2-month continuous monitoring data of egrets, our results cover the findings using conventional manual observations, e.g., vertical stratification of egrets according to body size, and also open up opportunities of long-term bird surveys requiring intensive monitoring that is impractical using conventional methods, e.g., the weather influences on egrets, and the relationship of the migration schedules between the great egrets and little egrets.

scholarly journals Long-Term Data Traffic Forecasting for Network Dimensioning in LTE with Short Time Series

Electronics ◽  
2021 ◽  
Vol 10 (10) ◽  
pp. 1151
Author(s):  
Carolina Gijón ◽  
Matías Toril ◽  
Salvador Luna-Ramírez ◽  
María Luisa Marí-Altozano ◽  
José María Ruiz-Avilés
Keyword(s):  
Time Series ◽  
Deep Learning ◽  
Time Series Analysis ◽  
Data Traffic ◽  
Medium Term ◽  
Short Term ◽  
Traffic Forecasting ◽  
Series Analysis ◽  
Radio Planning

Network dimensioning is a critical task in current mobile networks, as any failure in this process leads to degraded user experience or unnecessary upgrades of network resources. For this purpose, radio planning tools often predict monthly busy-hour data traffic to detect capacity bottlenecks in advance. Supervised Learning (SL) arises as a promising solution to improve predictions obtained with legacy approaches. Previous works have shown that deep learning outperforms classical time series analysis when predicting data traffic in cellular networks in the short term (seconds/minutes) and medium term (hours/days) from long historical data series. However, long-term forecasting (several months horizon) performed in radio planning tools relies on short and noisy time series, thus requiring a separate analysis. In this work, we present the first study comparing SL and time series analysis approaches to predict monthly busy-hour data traffic on a cell basis in a live LTE network. To this end, an extensive dataset is collected, comprising data traffic per cell for a whole country during 30 months. The considered methods include Random Forest, different Neural Networks, Support Vector Regression, Seasonal Auto Regressive Integrated Moving Average and Additive Holt–Winters. Results show that SL models outperform time series approaches, while reducing data storage capacity requirements. More importantly, unlike in short-term and medium-term traffic forecasting, non-deep SL approaches are competitive with deep learning while being more computationally efficient.

Real-time multiple object tracking using deep learning methods

2021 ◽  
Author(s):  
Dimitrios Meimetis ◽  
Ioannis Daramouskas ◽  
Isidoros Perikos ◽  
Ioannis Hatzilygeroudis
Keyword(s):  
Deep Learning ◽  
Object Tracking ◽  
Real Time ◽  
Multiple Object Tracking ◽  
Learning Methods ◽  
Multiple Object

scholarly journals Multiple Pedestrians and Vehicles Tracking in Aerial Imagery Using a Convolutional Neural Network

2021 ◽  
Vol 13 (10) ◽  
pp. 1953
Author(s):  
Seyed Majid Azimi ◽  
Maximilian Kraus ◽  
Reza Bahmanyar ◽  
Peter Reinartz
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Convolutional Neural Network ◽  
Object Tracking ◽  
Short Term Memory ◽  
Aerial Imagery ◽  
Future Research ◽  
Short Term ◽  
Term Memory ◽  
Long Short Term Memory

In this paper, we address various challenges in multi-pedestrian and vehicle tracking in high-resolution aerial imagery by intensive evaluation of a number of traditional and Deep Learning based Single- and Multi-Object Tracking methods. We also describe our proposed Deep Learning based Multi-Object Tracking method AerialMPTNet that fuses appearance, temporal, and graphical information using a Siamese Neural Network, a Long Short-Term Memory, and a Graph Convolutional Neural Network module for more accurate and stable tracking. Moreover, we investigate the influence of the Squeeze-and-Excitation layers and Online Hard Example Mining on the performance of AerialMPTNet. To the best of our knowledge, we are the first to use these two for regression-based Multi-Object Tracking. Additionally, we studied and compared the L1 and Huber loss functions. In our experiments, we extensively evaluate AerialMPTNet on three aerial Multi-Object Tracking datasets, namely AerialMPT and KIT AIS pedestrian and vehicle datasets. Qualitative and quantitative results show that AerialMPTNet outperforms all previous methods for the pedestrian datasets and achieves competitive results for the vehicle dataset. In addition, Long Short-Term Memory and Graph Convolutional Neural Network modules enhance the tracking performance. Moreover, using Squeeze-and-Excitation and Online Hard Example Mining significantly helps for some cases while degrades the results for other cases. In addition, according to the results, L1 yields better results with respect to Huber loss for most of the scenarios. The presented results provide a deep insight into challenges and opportunities of the aerial Multi-Object Tracking domain, paving the way for future research.

High speed long-term visual object tracking algorithm for real robot systems

2021 ◽  
Vol 434 ◽  
pp. 268-284
Author(s):  
Muxi Jiang ◽  
Rui Li ◽  
Qisheng Liu ◽  
Yingjing Shi ◽  
Esteban Tlelo-Cuautle
Keyword(s):  
Object Tracking ◽  
High Speed ◽  
Tracking Algorithm ◽  
Visual Object ◽  
Visual Object Tracking ◽  
Robot Systems ◽  
Real Robot

scholarly journals Deep Learning of Appearance Affinity for Multi-Object Tracking and Re-Identification: A Comparative View

Electronics ◽  
2020 ◽  
Vol 9 (11) ◽  
pp. 1757
Author(s):  
María J. Gómez-Silva ◽  
Arturo de la Escalera ◽  
José M. Armingol
Keyword(s):  
Deep Learning ◽  
Object Tracking ◽  
Loss Function ◽  
Neural Model ◽  
Training Data ◽  
Learning Approaches ◽  
The Core ◽  
Triplet Loss ◽  
Affinity Model

Recognizing the identity of a query individual in a surveillance sequence is the core of Multi-Object Tracking (MOT) and Re-Identification (Re-Id) algorithms. Both tasks can be addressed by measuring the appearance affinity between people observations with a deep neural model. Nevertheless, the differences in their specifications and, consequently, in the characteristics and constraints of the available training data for each one of these tasks, arise from the necessity of employing different learning approaches to attain each one of them. This article offers a comparative view of the Double-Margin-Contrastive and the Triplet loss function, and analyzes the benefits and drawbacks of applying each one of them to learn an Appearance Affinity model for Tracking and Re-Identification. A batch of experiments have been conducted, and their results support the hypothesis concluded from the presented study: Triplet loss function is more effective than the Contrastive one when an Re-Id model is learnt, and, conversely, in the MOT domain, the Contrastive loss can better discriminate between pairs of images rendering the same person or not.

THE POWER OF TESTS: A CRITICAL PERSPECTIVE ON THE USES OF LANGUAGE TESTS.Elana Shohamy. London: Longman, 2001. Pp. xxvi + 182. £18.99 paper.

2002 ◽  
Vol 24 (4) ◽  
pp. 644-645 ◽  
Author(s):  
April Ginther
Keyword(s):  
School Administrators ◽  
Educational Institutions ◽  
Motivating Factors ◽  
Critical Perspective ◽  
Policy Makers ◽  
Power Of Tests ◽  
Language Tests ◽  
Short And Long Term ◽  
Long Term Consequences

In the introduction to The power of tests: A critical perspective on the uses of language tests, Elana Shohamy raises the following questions: What is the meaning of a test for test takers, parents, teachers, and school administrators? What are the short- and long-term consequences of tests on the lives of individuals? What are the motivating factors behind the administration of language tests? What are the politics of the tests? These kinds of questions logically arise when the examination of testing includes a concern with the use of tests by educational institutions, policy makers, and society at large. Focusing primarily on the misuse of tests, this volume chronicles both intended and unintended test consequences.

Spoken content retrieval and understanding using deep learning

Impact ◽  
2021 ◽  
Vol 2021 (1) ◽  
pp. 9-11
Author(s):  
Lin-shan Lee
Keyword(s):  
Deep Learning ◽  
Scientific Community ◽  
Online Courses ◽  
The Internet ◽  
Human Voice ◽  
Content Retrieval ◽  
Youtube Videos ◽  
Access Information ◽  
Teaching Machines

Spoken content refers to all content over the Internet which includes human voice, essentially those in multimedia, such As YouTube videos and online courses. Today such content is retrieved via Google primarily based on human-generated text labels, because Google can only retrieve text over the Internet. The goal of this project is to produce technologies to retrieve accurately and efficiently such spoken content directly based on the included audio sounds instead of text labels, because machines today can listen to human voice just as they can read the text. The long term goal is to create a spoken version of Google, which may revolutionize the ways in which humans access information and improve their knowledge. Professor Lin-shan Lee at National Taiwan University is leading this project. He has been a distinguished leader in the global scientific community for the area of teaching machines to speak and listen to human voice for many years.

Sign in / Sign up

Export Citation Format

Share Document