scholarly journals A SURVEY OF MULTISPECTRAL IMAGE DENOISING METHODS FOR SATELLITE IMAGERY APPLICATIONS

2017 ◽  
Vol 10 (13) ◽  
pp. 292
Author(s):  
Ankush Rai ◽  
Jagadeesh Kannan R
Keyword(s):  
Performance Measures ◽  
Learning Algorithm ◽  
Research Question ◽  
Object Extraction ◽  
Multispectral Images ◽  
High Definition ◽  
Learned Dictionary ◽  
Object Tagging ◽  
The Given ◽  
Image Object

In comparison with the standard RGB or gray-scale images, the usual multispectral images (MSI) are intended to convey high definition and anauthentic representation for real world scenes to significantly enhance the performance measures of several other tasks involving with computervision, segmentation of image, object extraction, and object tagging operations. While procuring images form satellite, the MSI are often prone tonoises. Finding a good mathematical description of the learning-based denoising model is a difficult research question and many different researchesaccounted in the literature. Many have attempted its use with the application of neural network as a sparse learned dictionary of noisy patches.Furthermore, this approach allows several algorithm to optimize itself for the given task at hand using machine learning algorithm. However, inpractices, a MSI image is always prone to corruption by various sources of noises while procuring the images. In this survey, we studied the pasttechniques attempted for the noise influenced MSI images. The survey presents the outline of past techniques and their respective advantages incomparison with each other.

scholarly journals BI-DIRECTIONAL RECURRENT NEURAL NETWORK FOR IMPROVING MULTISPECTRAL IMAGE DENOISING

2017 ◽  
Vol 10 (13) ◽  
pp. 272
Author(s):  
Ankush Rai ◽  
Jagadeesh Kannan R
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Recurrent Neural Network ◽  
Learning Algorithm ◽  
Research Question ◽  
Machine Learning Algorithm ◽  
Multispectral Images ◽  
Improved Method ◽  
Learned Dictionary ◽  
The Given

While procuring images form satellite the multispectral images (MSI) are often prone to noises. finding a good mathematical description of the learning based denoising model is a difficult research question and many different research accounted in the literature. Many have attempted its use with the application of neural network as a sparse learned dictionary of noisy patches. Also, this approach allows several algorithm to optimize itself for the given task at hand by using machine learning algorithm. In this study we present an improved method for learning based denoising of MSI images. Recurrent neural network used in this study helps in speeding up the computational operability and denoising performance by over 85% to 95%.     

scholarly journals How Important is Gambling in National GDP: Case Study from Austria, Croatia, Italy and Slovenia

ECONOMICS ◽  
2019 ◽  
Vol 7 (1) ◽  
pp. 31-49
Author(s):  
Andrej Raspor ◽  
Iva Bulatović ◽  
Ana Stranjančević ◽  
Darko Lacmanović
Keyword(s):  
Research Question ◽  
The State ◽  
State Budget ◽  
Significant Information ◽  
Main Research ◽  
Benchmark Analysis ◽  
The World ◽  
State Budgets ◽  
The Given ◽  
First Time

Abstract Purpose – The situation in the field of gambling is changing due to the rise of Internet and Mobile gambling. In general gambling consumption is increasing every year, but the distribution of consumption has radically changed from Land Based gambling to Remote gambling. The purpose of this article is to present an overview of the world gambling industry and a specific overview in Austria, Croatia, Italy and Slovenia in order to find some main similarities and differences in observed period. Design/Methodology/Approach – The main research question is How important is gambling for the involved countries and what proportion of the national GDP does the gambling revenue account for? This paper presents the analysis of five statistical databases for the last sixteen years in order to find out some patterns, cyclical or seasonal features or other significant information that allows us to do forecasting of the future revenue with a certain degree of accuracy. We have systematically searched and collected data from the World Bank and the National Statistical Offices websites of the given countries. Statistical methods were used for benchmark analysis, while Box and Jenkins approach and ARIMA modelling were used for forecasting. Findings – The smallest increase was recorded in Slovenia and the largest in Italy. The same effects were also observed in the GDP of these countries. Thus, the state budgets of Croatia and Italy are increasingly dependent on gambling taxes. It also has negative wages. The gambling addictions among the locals have become more frequent as well. Originality of the research – The article shows the forecasts of the gambling revenue and its share in the GDP by 2027. We want to alert decision makers to adopt appropriate policies. States need to rethink their views on gambling and the excessive dependence of the state budget on gambling taxes. This is the first time a single comparative analysis of these countries and the above mentioned forecast has been conducted.

scholarly journals Parametric Methods to Multispectral Image Classification using Normalized Difference Vegetation Index

2019 ◽  
Vol 9 (2S) ◽  
pp. 611-615
Keyword(s):  
Land Cover ◽  
Vegetation Index ◽  
Normalized Difference Vegetation Index ◽  
Learning Algorithm ◽  
Urban Region ◽  
Landsat 8 ◽  
Multispectral Images ◽  
Barren Soil ◽  
Day By Day ◽  
Manual Verification

The key to proper governance of the municipal bodies lies in knowing the geography of the region. The land cover of the region changes with respect to time. Also, there are seasonal variation in the layout of the waterbodies. Manual verification and surveying of these things becomes very difficult for want of resources. Remote Sensing Images play a very important role in mapping the land cover. In this paper, we consider such remotely sensed Multispectral Images, taken from Landsat-8. Parametric Machine learning algorithm like Maximum Likelihood Classifier has been used on those images to classify the land cover. Normalized Difference Vegetation Index (NDVI) has been calculated and integrates with the classification process. Four basic land covers have been identified for the purpose namely Water, Vegetation, Built-up and Barren soil. The area of study is Bangalore urban region where we find that the water bodies are decreasing day by day. An overall efficiency of 82% with a kappa hat 0f 0.67 has been achieved with the method. The user and the producer accuracies have also been tabulated in the Results part. The results show the land cover changes in a temporal manner

CHEMICAL EQUILIBRIUM IN TERMS OF ITS CONCEPTUAL UNDERSTANDING IN THE CONTEXT OF SUBMICROSCOPIC, MACROSCOPIC AND SYMBOLIC INTERPRETATION BY LEARNERS

2017 ◽  
Author(s):  
Miroslav Prokša ◽  
◽  
Zuzana Haláková ◽  
Anna Drozdíková ◽  
◽  
...  
Keyword(s):  
Conceptual Understanding ◽  
Chemical Equilibrium ◽  
Research Question ◽  
Research Tool ◽  
Symbolic Representations ◽  
Symbolic Interpretation ◽  
The Given

The research was focused on solving the following research question: What is the depth and breadth of 16-year-old learners' knowledge of the chemical equilibrium in Slovakia? The main aim of our research was to find out the conceptual understanding of this part of chemistry in the context of submicroscopic, macroscopic and symbolic representations. A special research tool, which consisted of five sets of tasks, was created for this research. The research included a sample of 473 children. The results indicate that knowledge is more at the level of memory reproduction and algorithmic use. Learners have been facing a problem with the conceptual understanding of the given concept. Keywords: chemical equilibrium, submicroscopic, macroscopic and symbolic interpretation, conceptual understanding.

Evaluation of Remote Interface Component Alternatives for Teaching Tele-Robotic Operation

2012 ◽  
pp. 163-184
Author(s):  
Goldstain Ofir ◽  
Ben-Gal Irad ◽  
Bukchin Yossi
Keyword(s):  
Performance Measures ◽  
End Users ◽  
Computer Integrated Manufacturing ◽  
Learning Ability ◽  
Improvement Rate ◽  
Tel Aviv ◽  
Hands On ◽  
Overall Performance ◽  
The Given ◽  
Remote Learning

This chapter discusses a remote learning study conducted at the Computer-Integrated-Manufacturing (CIM) Laboratory in Tel-Aviv University. The goal is to provide remote end-users with an interface that enables them to teleoperate a robotic arm in conditions as close as possible to hands-on operation in the laboratory. This study evaluates the contribution of different interface components to the overall performance and the learning ability of potential end-users. Based on predefined experimental tasks, the study compares alternative interface designs for teleoperation. The three performance measures of the robot operation task are (1) the number of steps that are required to complete the given task, (2) the number of errors during the execution stage, and (3) the improvement rate of users. Guidelines for a better design of remote learning interfaces in robotics are provided based on the experimental results.

scholarly journals Enhanced Meta-Learning for Cross-Lingual Named Entity Recognition with Minimal Resources

2020 ◽  
Vol 34 (05) ◽  
pp. 9274-9281
Author(s):  
Qianhui Wu ◽  
Zijia Lin ◽  
Guoxin Wang ◽  
Hui Chen ◽  
Börje F. Karlsson ◽  
...  
Keyword(s):  
Learning Algorithm ◽  
Named Entity Recognition ◽  
Entity Recognition ◽  
Target Language ◽  
Test Case ◽  
Named Entity ◽  
Meta Learning ◽  
Target Languages ◽  
Cross Lingual ◽  
The Given

For languages with no annotated resources, transferring knowledge from rich-resource languages is an effective solution for named entity recognition (NER). While all existing methods directly transfer from source-learned model to a target language, in this paper, we propose to fine-tune the learned model with a few similar examples given a test case, which could benefit the prediction by leveraging the structural and semantic information conveyed in such similar examples. To this end, we present a meta-learning algorithm to find a good model parameter initialization that could fast adapt to the given test case and propose to construct multiple pseudo-NER tasks for meta-training by computing sentence similarities. To further improve the model's generalization ability across different languages, we introduce a masking scheme and augment the loss function with an additional maximum term during meta-training. We conduct extensive experiments on cross-lingual named entity recognition with minimal resources over five target languages. The results show that our approach significantly outperforms existing state-of-the-art methods across the board.

ACCELERATED LEARNING BY ACTIVE EXAMPLE SELECTION

1994 ◽  
Vol 05 (01) ◽  
pp. 67-75 ◽  
Author(s):  
BYOUNG-TAK ZHANG
Keyword(s):  
Neural Networks ◽  
Gradient Descent ◽  
Learning Algorithm ◽  
Accelerated Learning ◽  
Training Set ◽  
Alternative Approach ◽  
Speed Up ◽  
Multilayer Neural Networks ◽  
Training Examples ◽  
The Given

Much previous work on training multilayer neural networks has attempted to speed up the backpropagation algorithm using more sophisticated weight modification rules, whereby all the given training examples are used in a random or predetermined sequence. In this paper we investigate an alternative approach in which the learning proceeds on an increasing number of selected training examples, starting with a small training set. We derive a measure of criticality of examples and present an incremental learning algorithm that uses this measure to select a critical subset of given examples for solving the particular task. Our experimental results suggest that the method can significantly improve training speed and generalization performance in many real applications of neural networks. This method can be used in conjunction with other variations of gradient descent algorithms.

Modeling The Tower Of Hanoi Using Neural Network

1993 ◽  
pp. 47-56
Author(s):  
Mohamed Othman ◽  
Mohd. Hassan Selamat ◽  
Zaiton Muda ◽  
Lili Norliya Abdullah
Keyword(s):  
Neural Network ◽  
Knowledge Representation ◽  
Network Architecture ◽  
Learning Algorithm ◽  
Learning Rule ◽  
Neural System ◽  
Neural Systems ◽  
Tower Of Hanoi ◽  
Artificial Neural ◽  
The Given

This paper discusses the modeling of Tower of Hanoi using the concepts of neural network. The basis idea of backpropagation learning algorithm in Artificial Neural Systems is then described. While similar in some ways, Artificial Neural System learning deviates from tradition in its dependence on the modification of individual weights to bring about changes in a knowledge representation distributed across connection in a network. This unique form of learning is analyzed from two aspects: the selection of an appropriate network architecture for representing the problem, and the choice of a suitable learning rule capable qf reproducing the desired function within the given network. Key words: Tower of Hanoi; Backpropagation Algorithm; Knowledge Representation;

scholarly journals An Approach to Data Reduction for Learning from Big Datasets: Integrating Stacking, Rotation, and Agent Population Learning Techniques

Complexity ◽  
2018 ◽  
Vol 2018 ◽  
pp. 1-13 ◽  
Author(s):  
Ireneusz Czarnowski ◽  
Piotr Jędrzejowicz
Keyword(s):  
Data Reduction ◽  
Learning Algorithm ◽  
Research Question ◽  
Principal Component ◽  
Component Analysis ◽  
Classifier Ensembles ◽  
Agent Based ◽  
Reduction Techniques ◽  
Learning Techniques ◽  
Population Learning

In the paper, several data reduction techniques for machine learning from big datasets are discussed and evaluated. The discussed approach focuses on combining several techniques including stacking, rotation, and data reduction aimed at improving the performance of the machine classification. Stacking is seen as the technique allowing to take advantage of the multiple classification models. The rotation-based techniques are used to increase the heterogeneity of the stacking ensembles. Data reduction makes it possible to classify instances belonging to big datasets. We propose to use an agent-based population learning algorithm for data reduction in the feature and instance dimensions. For diversification of the classifier ensembles within the rotation also, alternatively, principal component analysis and independent component analysis are used. The research question addressed in the paper is formulated as follows: does the performance of a classifier using the reduced dataset be improved by integrating the data reduction mechanism with the rotation-based technique and the stacking?

Sign in / Sign up

Export Citation Format

Share Document