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

Saini, Rashmi

doi:10.7494/geom.2023.17.1.57

Cited by 11 publications

(10 citation statements)

References 26 publications

(78 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this study, the NIR had a lower contribution to classification accuracy than the other bands mentioned above, despite the fact that its important role in vegetation mapping is well known and proven [7,90]. NIR plays a key role in satellites with a higher spatial but lower spectral resolution than Sentinel-2, such as IKONOS-2 and WorldView-2 [91,92].…”

Section: Formula Comparisonmentioning

confidence: 71%

Evaluation and Selection of Multi-Spectral Indices to Classify Vegetation Using Multivariate Functional Principal Component Analysis

Pesaresi,

Mancini,

Quattrini

et al. 2024

Remote Sensing

View full text Add to dashboard Cite

The identification, classification and mapping of different plant communities and habitats is of fundamental importance for defining biodiversity monitoring and conservation strategies. Today, the availability of high temporal, spatial and spectral data from remote sensing platforms provides dense time series over different spectral bands. In the case of supervised mapping, time series based on classical vegetation indices (e.g., NDVI, GNDVI, …) are usually input characteristics, but the selection of the best index or set of indices (which guarantees the best performance) is still based on human experience and is also influenced by the study area. In this work, several different time series, based on Sentinel-2 images, were created exploring new combinations of bands that extend the classic basic formulas as the normalized difference index. Multivariate Functional Principal Component Analysis (MFPCA) was used to contemporarily decompose the multiple time series. The principal multivariate seasonal spectral variations identified (MFPCA scores) were classified by using a Random Forest (RF) model. The MFPCA and RF classifications were nested into a forward selection strategy to identify the proper and minimum set of indices’ (dense) time series that produced the most accurate supervised classification of plant communities and habitat. The results we obtained can be summarized as follows: (i) the selection of the best set of time series is specific to the study area and the habitats involved; (ii) well-known and widely used indices such as the NDVI are not selected as the indices with the best performance; instead, time series based on original indices (in terms of formula or combination of bands) or underused indices (such as those derivable with the visible bands) are selected; (iii) MFPCA efficiently reduces the dimensionality of the data (multiple dense time series) providing ecologically interpretable results representing an important tool for habitat modelling outperforming conventional approaches that consider only discrete time series.

show abstract

Section: Formula Comparisonmentioning

confidence: 71%

Evaluation and Selection of Multi-Spectral Indices to Classify Vegetation Using Multivariate Functional Principal Component Analysis

Pesaresi,

Mancini,

Quattrini

et al. 2024

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…The theoretical and practical importance is stimulating the development of this scientific direction. However, in this field, the Braun-Blanquet approach is clearly inferior to other scientific approaches based on remote sensing of territories [123][124][125][126].…”

Section: Clustermentioning

confidence: 92%

Global Overview of the Application of the Braun-Blanquet Approach in Research

Ivanova

2024

Forests

View full text Add to dashboard Cite

Environmental classifications are of paramount importance for assessing the impacts of land-use changes, for prioritizing conservation efforts, and for developing effective management strategies to mitigate the negative impacts of human activities. The aim of our research was to provide as complete an analysis as possible of the studies that have been carried out using the Braun-Blanquet approach. The global review of studies based on the Braun-Blanquet approach includes 1168 papers and was conducted using the PRISMA 2009 methodological recommendations, strict criteria for the selection/quality of papers, and modern methods of data analysis and visualization using VOSviewer software developed by Nees Jan van Eck and Ludo Waltman (Centre for Science and Technology Studies (CWTS) of Leiden University in the Netherlands) (version 1.6.18), which ensures a representative sample, minimization of subjective judgements, and reliability of conclusions. It was noted that the number of publications on Braun-Blanquet is growing exponentially. This is an indication of the scientific interest in this methodology and its continuous further development. Based on a detailed analysis of the keywords, the main research directions and challenges are identified. These include improving the conceptual and methodological foundations of the Braun-Blanquet approach; improvement in regional vegetation classifications, synthesizing them and producing a comprehensive classification for large areas as a basis for biodiversity conservation and sustainable land use; expansion of the geography; compilation and updating of databases of phytosociological data; management of dynamics and vegetation; discussion of the important problem of continuity and discreteness of vegetation in the context of ecological classifications; and vegetation mapping. The top 20 journals publishing the most cited articles were identified, as well as the top 20 most cited journals whose high citation rate is due to the large number of high-quality articles. The analysis of the bibliographic network of papers in dynamics has shown that the structure of relationships is not constant and has changed significantly. The analysis of the authors’ publication activity showed that the vast majority of researchers have a low publication activity and have published only one to three papers. A peculiarity also emerges: if all the most cited authors are concentrated in Eurasia, then most of the most actively published authors are outside Eurasia. The importance of the Braun-Blanquet approach for the study and classification of forest vegetation should be emphasized. In this case, the Braun-Blanquet approach is integrated into forest typologies, increasing their ecological validity and environmental relevance.

show abstract

“…AB is an ML classification algorithm; its principle is based on building strong classifiers by combining basic or weak classifiers. This classification algorithm works on adaptive sampling to select the between samples [40]. This algorithm iteratively trains the weak classifiers, for which it uses weighted data to incorporate it into an ensemble, to then have the strong classifier [41], as shown in Fig.…”

Section: Adaboostmentioning

confidence: 99%

Breast Cancer Prediction using Machine Learning Models

Iparraguirre-Villanueva¹,

Epifanía-Huerta²,

Torres-Ceclén³

et al. 2023

IJACSA

View full text Add to dashboard Cite

Breast cancer is a type of cancer that develops in the cells of the breast. Treatment for breast cancer usually involves X-ray, chemotherapy, or a combination of both treatments. Detecting cancer at an early stage can save a person's life. Artificial intelligence (AI) plays a very important role in this area. Therefore, predicting breast cancer remains a very challenging issue for clinicians and researchers. This work aims to predict the probability of breast cancer in patients. Using machine learning (ML) models such as Multilayer Perceptron (MLP), K-Nearest Neightbot (KNN), AdaBoost (AB), Bagging, Gradient Boosting (GB), and Random Forest (RF). The breast cancer diagnostic medical dataset from the Wisconsin repository has been used. The dataset includes 569 observations and 32 features. Following the data analysis methodology, data cleaning, exploratory analysis, training, testing, and validation were performed. The performance of the models was evaluated with the parameters: classification accuracy, specificity, sensitivity, F1 count, and precision. The training and results indicate that the six trained models can provide optimal classification and prediction results. The RF, GB, and AB models achieved 100% accuracy, outperforming the other models. Therefore, the suggested models for breast cancer identification, classification, and prediction are RF, GB, and AB. Likewise, the Bagging, KNN, and MLP models achieved a performance of 99.56%, 95.82%, and 96.92%, respectively. Similarly, the last three models achieved an optimal yield close to 100%. Finally, the results show a clear advantage of the RF, GB, and AB models, as they achieve more accurate results in breast cancer prediction.

show abstract

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

Cited by 11 publications

References 26 publications

Evaluation and Selection of Multi-Spectral Indices to Classify Vegetation Using Multivariate Functional Principal Component Analysis

Evaluation and Selection of Multi-Spectral Indices to Classify Vegetation Using Multivariate Functional Principal Component Analysis

Global Overview of the Application of the Braun-Blanquet Approach in Research

Breast Cancer Prediction using Machine Learning Models

Contact Info

Product

Resources

About