No abstract

Recent imaging science and technology discoveries have considered hyperspectral imagery and remote sensing. The current intelligent technologies, such as support vector machines, sparse representations, active learning, extreme learning machines, transfer learning, and deep learning, are typically based on the learning of the machines. These techniques enrich the processing of such three-dimensional, multiple bands, and high-resolution images with their precision and fidelity. This article presents an extensive survey depicting machine-dependent technologies’ contributions and deep learning on landcover classification based on hyperspectral images. The objective of this study is three-fold. First, after reading a large pool of Web of Science (WoS), Scopus, SCI, and SCIE-indexed and SCIE-related articles, we provide a novel approach for review work that is entirely systematic and aids in the inspiration of finding research gaps and developing embedded questions. Second, we emphasize contemporary advances in machine learning (ML) methods for identifying hyperspectral images, with a brief, organized overview and a thorough assessment of the literature involved. Finally, we draw the conclusions to assist researchers in expanding their understanding of the relationship between machine learning and hyperspectral images for future research.

Hyperspectral image (HSI) analysis generally suffers from issues such as high dimensionality, imbalanced sample sets for different classes, and the choice of classifiers for artificially balanced datasets. The existing conventional data imbalance removal techniques and forest classifiers lack a more efficient approach to dealing with the aforementioned issues. In this study, we propose a novel hybrid methodology ADASYN-enhanced subsampled multi-grained cascade forest (ADA-Es-gcForest) which comprises four folds: First, we extracted the most discriminative global spectral features by reducing the vast dimensions, i.e., the redundant bands using principal component analysis (PCA). Second, we applied the subsampling-based adaptive synthetic minority oversampling method (ADASYN) to augment and balance the dataset. Third, we used the subsampled multi-grained scanning (Mg-sc) to extract the minute local spatial–spectral features by adaptively creating windows of various sizes. Here, we used two different forests—a random forest (RF) and a complete random forest (CRF)—to generate the input joint-feature vectors of different dimensions. Finally, for classification, we used the enhanced deep cascaded forest (CF) that improvised in the dimension reduction of the feature vectors and increased the connectivity of the information exchange between the forests at the different levels, which elevated the classifier model’s accuracy in predicting the exact class labels. Furthermore, the experiments were accomplished by collecting the three most appropriate, publicly available his landcover datasets—the Indian Pines (IP), Salinas Valley (SV), and Pavia University (PU). The proposed method achieved 91.47%, 98.76%, and 94.19% average accuracy scores for IP, SV, and PU datasets. The validity of the proposed methodology was testified against the contemporary state-of-the-art eminent tree-based ensembled methods, namely, RF, rotation forest (RoF), bagging, AdaBoost, extreme gradient boost, and deep multi-grained cascade forest (DgcForest), by simulating it numerically. Our proposed model achieved correspondingly higher accuracies than those classifiers taken for comparison for all the HS datasets.

Imbalance in hyperspectral images creates a crisis in its analysis and classification operation. Resampling techniques are utilized to minimize the data imbalance. Although only a limited number of resampling methods were explored in the previous research, a small quantity of work has been done. In this study, we propose a novel illustrative study of the performance of the existing resampling techniques, viz. oversampling, undersampling, and hybrid sampling, for removing the imbalance from the minor samples of the hyperspectral dataset. The balanced dataset is classified in the next step, using the tree-based ensemble classifiers by including the spectral and spatial features. Finally, the comparative study is performed based on the statistical analysis of the outcome obtained from those classifiers that are discussed in the results section. In addition, we applied a new ensemble hybrid classifier named random rotation forest to our dataset. Three benchmark hyperspectral datasets: Indian Pines, Salinas Valley, and Pavia University, are applied for performing the experiments. We have taken precision, recall, F score, Cohen kappa, and overall accuracy as assessment metrics to evaluate our model. The obtained result shows that SMOTE, Tomek Links, and their combinations stand out to be the more optimized resampling strategies. Moreover, the ensemble classifiers such as rotation forest and random rotation ensemble provide more accuracy than others of their kind.

No abstract