Development of a hybrid classification technique based on deep learning applied to MSG / SEVIRI multispectral data

certain piece of textual information produced by any user or agent is said to be inappropriate if the expressed intent can cause hate, annoyance to other users or exhibits lack of respect, rudeness, which is disrespectful towards certain individuals or communities who may cause harm to oneself or others. In the present day scenario the different classification techniques are used to filter this kind of annoying text or messages. And browsers this days should be able to filter such kind of searches done in the searching engines which will be done every day. Providing such classification technique to filter such messages or searches which are not appropriate using some of the deep learning algorithms and considering the web search conversations such kind of searches which is found as abusive or which might cause hatred can be eliminated.

Download Full-text

BENCHMARKING DEEP LEARNING FRAMEWORKS FOR THE CLASSIFICATION OF VERY HIGH RESOLUTION SATELLITE MULTISPECTRAL DATA

ISPRS Annals of Photogrammetry Remote Sensing and Spatial Information Sciences ◽

10.5194/isprsannals-iii-7-83-2016 ◽

2016 ◽

Vol III-7 ◽

pp. 83-88 ◽

Cited By ~ 24

Author(s):

M. Papadomanolaki ◽

M. Vakalopoulou ◽

S. Zagoruyko ◽

K. Karantzalos

Keyword(s):

Remote Sensing ◽

Deep Learning ◽

High Resolution ◽

Remote Sensing Data ◽

Multispectral Data ◽

Sensing Data ◽

Multispectral Remote Sensing ◽

High Level ◽

Learning Frameworks

In this paper we evaluated deep-learning frameworks based on Convolutional Neural Networks for the accurate classification of multispectral remote sensing data. Certain state-of-the-art models have been tested on the publicly available SAT-4 and SAT-6 high resolution satellite multispectral datasets. In particular, the performed benchmark included the AlexNet, AlexNet-small and VGG models which had been trained and applied to both datasets exploiting all the available spectral information. Deep Belief Networks, Autoencoders and other semi-supervised frameworks have been, also, compared. The high level features that were calculated from the tested models managed to classify the different land cover classes with significantly high accuracy rates i.e., above 99.9%. The experimental results demonstrate the great potentials of advanced deep-learning frameworks for the supervised classification of high resolution multispectral remote sensing data.

Download Full-text

Deep learning based root-soil segmentation from X-ray tomography

10.1101/071662 ◽

2016 ◽

Cited By ~ 8

Author(s):

Clement DOUARRE ◽

Richard SCHIELEIN ◽

Carole FRINDEL ◽

Stefan GERTH ◽

David ROUSSEAU

Keyword(s):

Deep Learning ◽

Plant Science ◽

Low Contrast ◽

X Ray ◽

Classification Technique ◽

Plant Sciences ◽

Science Problem ◽

Classical Image ◽

First Time ◽

Segmentation Problem

One of the most challenging computer vision problem in plant sciences is the segmentation of root and soil from X-ray tomography. So far, this has been addressed from classical image analysis methods. In this paper, we address this root/soil segmentation problem from X-ray tomography using a new deep learning classification technique. The robustness of this technique, tested for the first time on this plant science problem, is established with root/soil presenting a very low contrast in X-ray tomography. We also demonstrate the possibility to segment efficiently root from soil while learning on purely synthetic soil and root.

Download Full-text

A Hybrid Classification Approach Based on Decision Tree and Naïve Bays Methods

International Journal of Information Retrieval Research ◽

10.4018/ijirr.2014100104 ◽

2014 ◽

Vol 4 (4) ◽

pp. 61-72

Author(s):

Saed A. Muqasqas ◽

Qasem A. Al Radaideh ◽

Bilal A. Abul-Huda

Keyword(s):

Data Mining ◽

Decision Tree ◽

Classification Accuracy ◽

Classification Methods ◽

Hybrid Classifier ◽

Classification Approach ◽

Classification Technique ◽

Average Accuracy ◽

Proposed Model ◽

Hybrid Classification

Data classification as one of the main tasks of data mining has an important role in many fields. Classification techniques differ mainly in the accuracy of their models, which depends on the method adopted during the learning phase. Several researchers attempted to enhance the classification accuracy by combining different classification methods in the same learning process; resulting in a hybrid-based classifier. In this paper, the authors propose and build a hybrid classifier technique based on Naïve Bayes and C4.5 classifiers. The main goal of the proposed model is to reduce the complexity of the NBTree technique, which is a well known hybrid classification technique, and to improve the overall classification accuracy. Thirty six samples of UCI datasets were used in evaluation. Results have shown that the proposed technique significantly outperforms the NBTree technique and some other classifiers proposed in the literature in term of classification accuracy. The proposed classification approach yields an overall average accuracy equal to 85.70% over the 36 datasets.

Download Full-text

Detection of Lung Malignancy Using SqueezeNet-Fc Deep Learning Classification Technique

Algorithms for Intelligent Systems - Proceedings of the International Conference on Paradigms of Communication, Computing and Data Sciences ◽

10.1007/978-981-16-5747-4_59 ◽

2022 ◽

pp. 683-699

Author(s):

Vinod Kumar ◽

Brijesh Bakariya

Keyword(s):

Deep Learning ◽

Classification Technique ◽

Lung Malignancy

Download Full-text

Machine learning for recognizing minerals from multispectral data

The Analyst ◽

10.1039/d0an01483d ◽

2021 ◽

Vol 146 (1) ◽

pp. 184-195

Author(s):

Pavel Jahoda ◽

Igor Drozdovskiy ◽

Samuel J. Payler ◽

Leonardo Turchi ◽

Loredana Bessone ◽

...

Keyword(s):

Machine Learning ◽

Deep Learning ◽

Data Fusion ◽

Learning Algorithms ◽

Multispectral Data

Combining Deep Learning algorithms, together with data fusion from multi-method spectroscopy, could drastically increase the accuracy of automatic mineral recognition compared to existing approaches.

Download Full-text