On Understanding Big Data Impacts in Remotely Sensed Image Classification Using Support Vector Machine Methods

Abstract

Abstract-Owing to the recent development of sensor resolutions on-board different Earth observation platforms, remote sensing is an important source of information for mapping and monitoring natural and man-made land covers. Of particular importance is the increasing amounts of available hyperspectral data originating from airborne and satellite sensors such as AVIRIS, HyMap, and Hyperion with very high spectral resolution (i.e., high number of spectral channels) containing rich information for a wide range of applications. A relevant example is the separation of different types of land cover classes using this data in order to understand for example impacts of natural disasters or changing of city buildings over time. More recently, such increases in the data volume, velocity, and variety of data contributed to the term "big data" that stands for challenges shared with many other scientific disciplines. On the one hand the amount of available data is increasing in a way that raises the demand for automatic data analysis elements since many of the available data collections are massively underutilized lacking experts for manual investigation. On the other hand, proven statistical methods (e.g.,, dimensionality reduction) driven by manual approaches have a significant impact in reducing the amount of "big data" towards smaller "smart data" contributing to the more recently used terms data value and veracity (i.e., less noise, lower dimensions that capture the most important information). This paper aims to take stock of which proven statistical data mining methods in remote sensing are used to contribute to "smart data" analysis processes in the light of possible automation as well as scalable and parallel processing techniques. We focus on parallel support vector machines (SVMs) as one of the best out-of-the-box classification methods.

Benchmarking Deep Learning Frameworks for the Classification of Very High Resolution Satellite Multispectral Data

“…Regarding automation, although unsupervised and semi-supervised approaches possess a native advantage, when comes to big data from space with important spatial, spectral and temporal variability, efficient generic tools may be based on supervised approaches which have been trained to handle and classify such datasets , Cavallaro et al, 2015.…”

A Novel Integrated Framework to Ensure Better Data Quality in Big Data Analytics over Cloud Environment

“…As the proposed system is nearly similar to optimize the performance of a framework for big data analytics, hence, it is wise enough to be compared with similar frequently used techniquesof optimization. We find that adoption of Support Vector Machine has been carried out by Cavallaro et al [20] as well as by Singh et al [24]. One of the optimization feature of SVM is its excellent classification approach that has capability to be applied over high-dimensioanl data and it is independent of any form of conventional feature selection procedures in over to resists the problem associated with larger dimensionality of big data.…”

“…The study outcome witness around 90% accuracy with 2.5% of error in classification performance. Cavallaro et al [20] have used learning algorithm for performing an effective classification of images. Lu et al [21] have presented a modeling of a concept that allows performing big data analytics over the cloud environment.…”

Comprehensive survey of deep learning in remote sensing: theories, tools, and challenges for the community

“…Provides overview of probabilistic models (undirected graphical, RBM, AE, SAE, DAE, contractive autoencoders, manifold learning, difficulty in training deep networks, handling high-dimensional inputs, evaluating performance, etc.) 330 Examines big-data impacts on SVM machine learning. 1 Covers about 170 publications in the area of scene classification and discusses limitations of datasets and problems associated with high-resolution imagery.…”