Image Retrieval Based on Self-Organizing Feature Map and Multilayer Perceptron Neural Networks Classifier

Ghaleb, Moshira S.; Ebied, Hala M.; Shedeed, Howida A.; Tolba, Mohamed F.

doi:10.1109/icicis46948.2019.9014768

Cited by 12 publications

(4 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Traditionally speaking, utilizing cloud coverage typically entails dispatching staff to the field to take ground-level photos during the observation stage, having knowledgeable staff analyses for images while in stations for the analysis phase [17], and then computing the desired outcome in a weather prediction using all additional information mentioned before, encompassing clouds, during the prediction phase. It is acknowledged on a global scale that a traditional meteorological institution may make forecasts with varied levels of approximation accuracy for particular time frames: 90% for a period of five days, 80% for a period of 50% after seven days for a period of 10 days [18]. Analysis for prediction tools and methodologies are still far from error-free, especially over longer time periods, despite the use of sophisticated technologies and procedures.…”

Section: Related Workmentioning

confidence: 99%

Using Deep learning to Improve the Classification of Weather Phenomena from Images

mohamad¹

2023

آفاق آسيوية

View full text Add to dashboard Cite

One of the most important phenomena at the moment is the identification of weather phenomena, which are crucial for numerous facets of daily life and are especially important for weather forecasts, transportation, tracking road conditions, farming, and forest management to protect of environment.Contrarily, few studies have attempted to classify images of actual weather events because it is challenging to do so using pictures, which are primarily reliant on human visual observations. This lessens the possibility of changing weather patterns. As far as we can determine, it takes time and is difficult to accurately distinguish between various weather events using traditional artificial vision. Even though some research has increased precision, also efficacy to recognizing phenomena employing artificial intelligence and even though AI approaches are better suited for categorization, they have discovered less different weather phenomena. This study suggests five artificial intelligence classification methods for meteorological events.In the meantime, we created a brand-The Weather Phenomenon Database (WEAPD), a brand-new dataset with 6,691 photos and 11 different types of weather phenomena, was just released. The categorization accuracy for AI methods in the WEAPD test group is about 84%. and the experimental results show how effective and better the models are for the specified AI State Information Service Issue 11 2023 12 Asian Perspectivestactics. Next studies on the classification of weather photographs and weather forecasting may use the creation of automatic, high-quality categorization of weather photos as a benchmark. The proportion is not higher than these values since the images are difficult to categorizes because they combine a variety of events.

show abstract

Section: Related Workmentioning

confidence: 99%

Using Deep learning to Improve the Classification of Weather Phenomena from Images

mohamad¹

2023

آفاق آسيوية

View full text Add to dashboard Cite

show abstract

“…Using machine learning to extract the important features from the images is the main factor in increasing the classification accuracy. Ghaleb et al [29] made a combined model of Selforganized feature map (SOFM) and Multilayer perceptron (MLP). The study has extracted the feature vector using SOFM.…”

Section: Related Workmentioning

confidence: 99%

Weather Classification using Fusion Of Convolutional Neural Networks and Traditional Classification Methods

Ghaleb

Moushier

Shedeed

et al. 2022

IJICIS

View full text Add to dashboard Cite

The weather phenomenon is very important in routine lives. The weather prediction, road electronic monitoring, traffic communication, capping inversion (CAP), afforestation, and the adjustment of the environmental issues are important factors to many decisions. Weather images classification may help in decision support systems. There are traditional and intelligent ways that can sufficiently achieve weather image classification. Traditional methods enhance the classification accuracy and the usability of weather phenomena. Researchers approve that machine learning has achieved better accuracies based on deep learning neural networks. This paper compares three different intelligent models by using a weather image dataset. The first model uses a convolution neural network (CNN) to classify five categories of weather images. The second model uses a fusion of convolution neural network and Decision Tree (DT). The third one uses a fusion of CNN and Support Vector Machine (SVM). The three models are applied to the collected dataset from Github and Kaggle. The study has achieved 92%, 93%, and 94% for CNN, CNN+DT, and CNN+SVM respectively. The Proposed methods have achieved high recognition accuracies for weather forecasting.

show abstract

“…Но обычно, данные алгоритмы используются как часть более сложной системы. К примеру алгоритм K-Means в работах [10][11] используется совместно с генетическим алгоритмом, а сеть Кохонена в [12][13][14] с персептроном.…”

Section: гибридные системыunclassified

Comparison of G-Means Algorithms and Kohonen Network in Solving Clustering Problems

Sudakov

Дмитриев

2022

Proceedings of the 32nd International Conference on Computer Graphics and Vision

View full text Add to dashboard Cite

Purpose: In this paper, the question of how to improve a self-organizing neural network consisting of a bundle of clustering algorithm and a multilayer perceptron for data verification tasks in the absence of training pairs is considered. Design/methodology/approach: The most popular clustering algorithm is the Kohonen network, but today it is not the only algorithm capable of performing the task quickly and accurately. The paper compares the Kohonen network and the G-Means algorithm. The principle of operation of these two algorithms is briefly analyzed. The accuracy of these algorithms and the speed of their learning are compared. Findings: By conducting experiments, conclusions were drawn about the speed and accuracy of the algorithms. Originality/value: The relevance of this work lies in the fact that the preparation of training pairs for intelligent systems and the process of learning with a teacher is a resource-intensive task. The systems of self-learning algorithms under consideration will significantly increase the learning rate, as well as eliminate the need for manual classification of data and the creation of training pairs for the perceptron, which in turn will allow you to create a self-learning system with the ability to generalize and predict.

show abstract

Image Retrieval Based on Self-Organizing Feature Map and Multilayer Perceptron Neural Networks Classifier

Cited by 12 publications

References 20 publications

Using Deep learning to Improve the Classification of Weather Phenomena from Images

Using Deep learning to Improve the Classification of Weather Phenomena from Images

Weather Classification using Fusion Of Convolutional Neural Networks and Traditional Classification Methods

Comparison of G-Means Algorithms and Kohonen Network in Solving Clustering Problems

Contact Info

Product

Resources

About