One of the main steps in hyperspectral image classification is the selection of bands that provide the best separability among classes. It is usually understood that the selected bands for classification must contain a large amount of information, and the correlation among selected bands should be small to avoid redundancy. At the same time for optimal classification, class separability should be at maximum value. The question arises whether the most informative spectral regions are really the same as the most discriminant ones for a given set of classes. Answering the question, we developed a new method named Spectral Region Splitting (SRS) to identify interesting spectral regions. This article concludes that the optimal informative and the optimal separable spectral regions are not identical. Furthermore, the cause of the difference is proven theoretically.@en

Hyperspectral images may be applied to classify objects in a scene. The redundancy in hyperspectral data implies that fewer spectral features might be sufficient for discriminating the objects captured in a scene. The availability of labeled classes of several areas in a scene paves the way for a supervised dimensionality reduction, i.e., using a discrimination measure between the classes in a scene to select spectral features. We show that averaging adjacent spectral channels and using wider spectral regions yield a better class separability than the selection of individual channels from the original hyperspectral dataset. We used a method named spectral region splitting (SRS), which creates a new feature space by averaging neighboring channels. In addition to the common benefits of channel selection methods, the algorithm constructs wider spectral regions when it is useful. Using different class separability measures over various datasets resulted in a better discrimination between the classes than the best-selected channels using the same measure. The reason is that the wider spectral regions led to a reduction in intraclass distances and an improvement in class discrimination. The overall classification accuracy of two hyperspectral scenes gave an increase of about two-percent when using the spectral regions determined by applying SRS. @en

This paper proposes a novel self-learned integrated framework of active learning (AL) and semi-supervised learning (SSL). SSL methods try to estimate a certain semilabel for unlabeled samples. While AL methods select the most informative unlabeled samples for the current classifying model and provide their labels by human expertise. An excessive human-machine interaction is required for labeling the selected instances. Whereas, providing reliable labels is a sensitive, time-consuming and expensive step. In our framework, we try to decrease the required human supervision by incorporating SSL method. In addition, the participation rate of each AL and SSL methods in the framework is adaptive and determined based on the certainty of the classifier at each iteration. The experiments were carried out on Pavia University image data which is an urban scene. The results showed the efficiency and the excellent performance of the proposed method in both terms of accuracy and computational cost.@en

This paper proposes a novel self-learned integrated framework of active learning (AL) and semi-supervised learning (SSL). SSL methods try to estimate a certain semilabel for unlabeled samples. While AL methods select the most informative unlabeled samples for the current classifying model and provide their labels by human expertise. An excessive human-machine interaction is required for labeling the selected instances. Whereas, providing reliable labels is a sensitive, time-consuming and expensive step. In our framework, we try to decrease the required human supervision by incorporating SSL method. In addition, the participation rate of each AL and SSL methods in the framework is adaptive and determined based on the certainty of the classifier at each iteration. The experiments were carried out on Pavia University image data which is an urban scene. The results showed the efficiency and the excellent performance of the proposed method in both terms of accuracy and computational cost.@en

Unsupervised feature selection (UFS) is a standard approach to reduce the dimensionality of hyperspectral images (HSIs). The main idea in UFS is to define a similarity metric, and select the features minimizing the metric to reduce the data redundancy. In this paper, we proposed a novel criterion for unsupervised dimensionality reduction based on the representation of spectral reflectance to capture dominant reflectance variations. Since capturing all the spectral information from an entire hyperspectral dataset is a time-consuming process, we proposed a heuristic algorithm named Greedy Search for Spectral Representation (GSSR). This algorithm divides the spectrum into spectral regions with less spectral variations and merges them. GSSR, similar to feature selection techniques, preserves the original data from being distorted or compromised by a transformation. We compared the GSSR algorithm with well-known existing algorithms in different experiments using various datasets. Comparison with the best approximation to represent single spectra as well as entire hyperspectral scene revealed that spectral representation is almost the same. The difference between the best spectral representation and the ones provided by GSSR is less than 0.01%; while on average, GSSR is about 660 times faster to represent single spectra and 37 times faster for a complete hyperspectral scene. Five well-known unsupervised dimensionality reduction methods were also implemented and used for comparison analysis. Based on the image classification accuracy over two hyperspectral datasets, the spectral features identified by the proposed criterion improved the classification accuracy as well.@en

Hyperspectral images present detailed spectral information of every pixel in the images where the spectral signal is sampled in hundreds of narrow and contiguous spectral channels, usually covering the 400-2500 nm spectral region where sunlight reflected by the Earth can be measured. Earth observation systems acquire spectral information by imaging spectrometers mounted in a platform flying over the Earth. Recent advances in technology make it possible to have miniaturised hyperspectral satellites in orbit. Much of the work presented in this thesis was inspired by the study of a CubeSat equipped with an imaging spectrometer and capable of onboard data processing.@en