Dimensionality Reduction and Classification of Hyperspectral Images Using Object-Based Image Analysis

Kavzoğlu, Taşkın; Tonbul, Hasan; Erdemir, Merve Yildiz; Çölkesen, İsmail

doi:10.1007/s12524-018-0803-1

Cited by 18 publications

(12 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The comparison demonstrated that all methods are very accurate (more than 90% of accuracy); however, the two-step method achieved the best results [147] Land change Proposed an unsupervised method with an OB approach to improve the detection of changes using high-resolution images. This methodology achieved better results in comparison to other methods [148] SGD 15 (Life on Land) Invasive plants Developed an unsupervised method to detect and map invasive plants using RFs, which proved to be a successful approach [149] Landslide Compared an unsupervised PB and OB approach for landslide detection using VHR images and concluded that OB performed better than PB [150] Land cover Compared four OB classifiers for the classification of a suburban area with data provided by Landsat-8 and proved that SVM had the best performance among all [151] Land use Proposed OB approach for urban land use classification using VHR images [152] SGD 15 (Life on Land) Land cover Tested the performance of PB and OB classification with a hyperspectral dataset and found that OB was better than PB approach [153] Land use Compared an OB and PB approach using aero photogrammetric images and the results showed that OB classifier performed better compared to PB [154] Disasters and Renewable Energy; the Regression category covers the SDGs 2, 3, 6, 7, 9, 11, 13, 14 and 15, and the fields Water Quality, Pollution and Freshwater; and the Dimension Reduction category covers the SDGs 3, 6, 7, 9, 11, 13 and 15, and the fields Land Cover, Electricity and Software. Thus, the overall findings confirm the significance of EO and ML in pursuing the goals of SD providing an overview of methods and techniques that sustain the achievement of SDGs.…”

Section: Discussionmentioning

confidence: 99%

Monitoring sustainable development by means of earth observation data and machine learning: a review

2020

View full text Add to dashboard Cite

This paper presents and explores the different Earth Observation approaches and their contribution to the achievement of United Nations Sustainable Development Goals. A review on the Sustainable Development concept and its goals is presented followed by Earth Observation approaches relevant to this field, giving special attention to the contribution of Machine Learning methods and algorithms as well as their potential and capabilities to support the achievement of Sustainable Development Goals. Overall, it is observed that Earth Observation plays a key role in monitoring the Sustainable Development Goals given its cost-effectiveness pertaining to data acquisition on all scales and information richness. Despite the success of Machine Learning upon Earth Observation data analysis, it is observed that performance is heavily dependent on the ability to extract and synthesise characteristics from data. Hence, a deeper and effective analysis of the available data is required to identify the strongest features and, hence, the key factors pertaining to Sustainable Development. Overall, this research provides a deeper understanding on the relation between Sustainable Development, Earth Observation and Machine Learning, and how these can support the Sustainable Development of countries and the means to find their correlations. In pursuing the Sustainable Development Goals, given the relevance and growing amount of data generated through Earth Observation, it is concluded that there is an increased need for new methods and techniques strongly suggesting the use of new Machine Learning techniques.

show abstract

Section: Discussionmentioning

confidence: 99%

Monitoring sustainable development by means of earth observation data and machine learning: a review

2020

View full text Add to dashboard Cite

show abstract

“…where L obj is the radiance of the recorded sample, L ref the radiance of the certified reflectance standard and ρ ref the reflectance of the certified reflectance standard (Peddle et al, 2001;Steffens and Buddenbaum, 2013) Hyperspectral images were processed in ENVI Classic (Version 5.2, Exelis Visual Information Solutions, Boulder, Colorado, United States) following the workflow in Figure 1. In particular, we used a principal component analysis (PCA) to concentrate the spectral information in images and remove the correlation between neighboring bands (Rodarmel and Shan, 2002;Kavzoglu et al, 2018). PCA images show spectral spatial variability of different components on the soil surface and helps us to define number of endmembers (classes) (Steffens et al, 2014).…”

Section: Imvnir Scanning At 53 µM Resolutionmentioning

confidence: 99%

Combination of Imaging Infrared Spectroscopy and X-ray Computed Microtomography for the Investigation of Bio- and Physicochemical Processes in Structured Soils

Lucas

Pihlap

Steffens

et al. 2020

Front. Environ. Sci.

View full text Add to dashboard Cite

Soil is a heterogeneous mixture of various organic and inorganic parent materials. Major soil functions are driven by their quality, quantity and spatial arrangement, resulting in soil structure. Physical protection of organic matter (OM) in this soil structure is considered as a vital mechanism for stabilizing organic carbon turnover, an important soil function in times of climate change. Herein, we present a technique for the correlative analysis of 2D imaging visible light near-infrared spectroscopy and 3D X-ray computed microtomography (µCT) to investigate the interplay of biogeochemical properties and soil structure in undisturbed soil samples. Samples from the same substrate but different soil management and depth (no-tilled topsoil, tilled topsoil and subsoil) were compared in order to evaluate this method in a diversely structured soil. Imaging spectroscopy is generally used to qualitatively and quantitatively identify OM with high spatial resolution, whereas 3D X-ray µCT provides high-resolution information on pore characteristics. The unique combination of these techniques revealed that, in undisturbed samples, OM can be found mainly at greater distances from macropores and close to biopores. However, alterations were observed because of disturbances by tillage. The correlative application of imaging infrared spectroscopic and X-ray µCT analysis provided new insights into the biochemical processes affected by soil structural changes.

show abstract

“…The ground reference dataset including 16 LULC classes was collected through a field study in June 1992 [36]. The Indian Pines dataset has been employed in many publications to test and compare the performances of various algorithms [37,38]. The dataset is regarded as a challenging one for classification problems because of three major reasons.…”

Section: The Indian Pines Datasetmentioning

confidence: 99%

Design of Feedforward Neural Networks in the Classification of Hyperspectral Imagery Using Superstructural Optimization

2020

Self Cite

View full text Add to dashboard Cite

Artificial Neural Networks (ANNs) have been used in a wide range of applications for complex datasets with their flexible mathematical architecture. The flexibility is favored by the introduction of a higher number of connections and variables, in general. However, over-parameterization of the ANN equations and the existence of redundant input variables usually result in poor test performance. This paper proposes a superstructure-based mixed-integer nonlinear programming method for optimal structural design including neuron number selection, pruning, and input selection for multilayer perceptron (MLP) ANNs. In addition, this method uses statistical measures such as the parameter covariance matrix in order to increase the test performance while permitting reduced training performance. The suggested approach was implemented on two public hyperspectral datasets (with 10% and 50% sampling ratios), namely Indian Pines and Pavia University, for the classification problem. The test results revealed promising performances compared to the standard fully connected neural networks in terms of the estimated overall and individual class accuracies. With the application of the proposed superstructural optimization, fully connected networks were pruned by over 60% in terms of the total number of connections, resulting in an increase of 4% for the 10% sampling ratio and a 1% decrease for the 50% sampling ratio. Moreover, over 20% of the spectral bands in the Indian Pines data and 30% in the Pavia University data were found statistically insignificant, and they were thus removed from the MLP networks. As a result, the proposed method was found effective in optimizing the architectural design with high generalization capabilities, particularly for fewer numbers of samples. The analysis of the eliminated spectral bands revealed that the proposed algorithm mostly removed the bands adjacent to the pre-eliminated noisy bands and highly correlated bands carrying similar information.

show abstract

Dimensionality Reduction and Classification of Hyperspectral Images Using Object-Based Image Analysis

Cited by 18 publications

References 30 publications

Monitoring sustainable development by means of earth observation data and machine learning: a review

Monitoring sustainable development by means of earth observation data and machine learning: a review

Combination of Imaging Infrared Spectroscopy and X-ray Computed Microtomography for the Investigation of Bio- and Physicochemical Processes in Structured Soils

Design of Feedforward Neural Networks in the Classification of Hyperspectral Imagery Using Superstructural Optimization

Contact Info

Product

Resources

About