Wyniki wyszukiwania - BazTech

1

Integrating Vegetation Indices and Spectral Features for Vegetation Mapping from Multispectral Satellite Imagery Using AdaBoost and Random Forest Machine Learning Classifiers

Saini Rashmi

Geomatics and Environmental Engineering

|

2023

|

Vol. 17, no. 1

57--74

EN

Vegetation mapping is an active research area in the domain of remote sensing. This study proposes a methodology for the mapping of vegetation by integrating several vegetation indices along with original spectral bands. The Land Use Land Cover classification was performed by two powerful Machine Learning techniques, namely Random Forest and AdaBoost. The Random Forest algorithm works on the concept of building multiple decision trees for the final prediction. The other Machine Learning technique selected for the classification is AdaBoost (adaptive boosting), converts a set of weak learners into strong learners. Here, multispectral satellite data of Dehradun, India, was utilised. The results demonstrate an increase of 3.87% and 4.32% after inclusion of selected vegetation indices by Random Forest and AdaBoost respectively. An Overall Accuracy (OA) of 91.23% (kappa value of 0.89) and 88.59% (kappa value of 0.86) was obtained by means of the Random Forest and AdaBoost classifiers respectively. Although Random Forest achieved greater OA as compared to AdaBoost, interestingly AdaBoost provided better class-specific accuracy for the Shrubland class compared to Random Forest. Furthermore, this study also evaluated the importance of each individual feature used in the classification. Results demonstrated that the NDRE, GNDVI, and RTVIcore vegetation indices, and spectral bands (NIR, and Red-Edge), obtained higher importance scores.

2

Entropy-based regularization of AdaBoost

Bereta M.

Computer Assisted Methods in Engineering and Science

|

2017

|

Vol. 24, no. 2

89--100

EN

In this study, we introduce an entropy-based method to regularize the AdaBoost algorithm. The AdaBoost algorithm is a well-known algorithm used to create aggregated classifiers. In many real-world classification problems in addition to paying special attention classification accuracy of the final classifier, great focus is placed on tuning the number of the so-called weak learners, which are aggregated by the final (strong)classifier. The proposed method is able to improve the AdaBoost algorithm in terms of both criteria. While many approaches to the regularization of boosting algorithms can be complicated, the proposed method is straightforward and easy to implement. We compare the results of the proposed method (EntropyAdaBoost) with the original AdaBoost and also with its regularized version, є-AdaBoost on several classification problems. It is shown that the proposed methods of EntropyAdaBoost and є-AdaBoost are strongly complementary when the improvement of AdaBoost is considered.

3

Automatic sleep scoring using statistical features in the EMD domain and ensemble methods

Hassan A. R., Bhuiyan M. I. H.

Biocybernetics and Biomedical Engineering

|

2016

|

Vol. 36, no. 1

248--255

EN

An automatic sleep scoring method based on single channel electroencephalogram (EEG) is essential not only for alleviating the burden of the clinicians of analyzing a high volume of data but also for making a low-power wearable sleep monitoring system feasible. However, most of the existing works are either multichannel or multiple physiological signal based or yield poor algorithmic performance. In this study, we propound a data-driven and robust automatic sleep staging scheme that uses single channel EEG signal. Decomposing the EEG signal segments using Empirical Mode Decomposition (EMD), we extract various statistical moment based features. The effectiveness of statistical features in the EMD domain is inspected. Statistical analysis is performed for feature selection. We then employ Adaptive Boosting and decision trees to perform classification. The performance of our feature extraction scheme is studied for various choices of classification models. Experimental outcomes manifest that the performance of the proposed sleep staging algorithm is better than that of the state-of-the-art ones. Furthermore, the proposed method's non-REM 1 stage detection accuracy is better than most of the existing works.

4

A novel approach for automatic detection and classification of suspicious lesions in breast ultrasound images

Karimi B., Krzyżak A.

Journal of Artificial Intelligence and Soft Computing Research

|

2013

|

Vol. 3, No. 4

265--276

EN

In this research, a new method for automatic detection and classification of suspected breast cancer lesions using ultrasound images is proposed. In this fully automated method, de-noising using fuzzy logic and correlation among ultrasound images taken from different angles is used. Feature selection using combination of sequential backward search, sequential forward search and distance-based methods is obtained. A new segmentation method based on automatic selection of seed points and region growing is proposed and classification of lesions into two malignant and benign classes using combination of AdaBoost, Artificial Neural Network and Fuzzy Support Vector Machine classifiers and majority voting is implemented.

5

Wpływ metody selekcji słów kluczowych na skuteczność klasyfikacji stron internetowych z wykorzystaniem algorytmu boostingu

Gąciarz T., Czajkowski K.

Czasopismo Techniczne. Nauki Podstawowe

|

2012

|

R. 109, z. 1-NP

61--73

PL

Artykuł porusza zagadnienia dotyczące klasyfikacji stron internetowych. Klasyfikacja przeprowadzana jest w oparciu o analizę struktury oraz zawartości stron. Pod uwagę brane są cechy zróżnicowanym charakterze, w tym między innymi cechy strukturalne, wizualne, tekstowe, łączy internetowych. Przy budowie klasyfikatorów wykorzystano algorytm AdaBoost. Skupiono się na wpływie metody selekcji słów kluczowych na skuteczność procesu klasyfikacji.

EN

The paper concerns the issues of web pages analysis process. The classification is performed based on the analysis of the structure as well content of pages. Various characteristics are taken into account including inter alia, structural, visual, text, web and links features. During the construction of classifiers the AdaBoost algorithm was applied. This paper focuses on the impact of keyword selection methods on the effectiveness of the classification process.