Extinction Profiles for the Classification of Remote Sensing Data

Abstract

With respect to recent advances in remote sensing technologies, the spatial resolution of airborne and spaceborne sensors is getting finer, which enables us to precisely analyze even small objects on the Earth. This fact has made the research area of developing efficient approaches to extract spatial and contextual information highly active. Among the existing approaches, morphological profile and attribute profile (AP) have gained great attention due to their ability to classify remote sensing data. This paper proposes a novel approach that makes it possible to precisely extract spatial and contextual information from remote sensing images. The proposed approach is based on extinction filters, which are used here for the first time in the remote sensing community. Then, the approach is carried out on two well-known high-resolution panchromatic data sets captured over Rome, Italy, and Reykjavik, Iceland. In order to prove the capabilities of the proposed approach, the obtained results are compared with the results from one of the strongest approaches in the literature, i.e., APs, using different points of view such as classification accuracies, simplification rate, and complexity analysis. Results indicate that the proposed approach can significantly outperform its alternative in terms of classification accuracies. In addition, based on our implementation, profiles can be generated in a very short processing time. It should be noted that the proposed approach is fully automatic. Index Terms-Attribute profile (AP), extinction profile (EP), image classification, random forest (RF), remote sensing data. I. INTRODUCTION U NDOUBTEDLY, supervised classification plays a fundamental role in the analysis of remote sensing images, and many applications, such as crop monitoring, forest applications, urban development, mapping and tracking, and risk management, can be handled by an efficient classifier [1].

Feature Profiles from Attribute Filtering for Classification of Remote Sensing Images

“…For example, with area attribute filtering, using standard deviation as output feature (FP σ ) can improve 3.2% of OA (84.73% compared to 81.53%) from the standard AP a with the same feature length of 21. For another comparison, the best performance of the recent extinction profiles [13] for this specific Reykjavik data set was 85.58% in OA and κ = 0.8145, which is inferior than our best performance.…”

“…For area attribute, ten thresholds were adopted for the Reykjavik data as proposed by several papers [5], [13]. For the Pavia University data, they were automatically computed according to the work in [16].…”

Assessing the performance of multiple spectral–spatial features of a hyperspectral image for classification of urban land cover classes using support vector machines and artificial neural network

“…They are proven to be effective in analysing and extracting spatial information that can consistently improve the classifier performance [20,35,42]. Although, MMs and their modifications (e.g., morphological profiles) have been used intensively in the remote sensing community, their concepts have a few limitations such as: (i) the shape of the SE is fixed, and consequently, they cannot efficiently extract spatial information; and (2) SEs are only able to extract information related to the size of existing objects, while they are unable to characterize information related to the gray-level characteristics of the regions [43].…”

Attribute Learning for SAR Image Classification

“…Model-based methods, e.g., Markov Random Fields (MRFs) [8], Conditional Random Fields [9], Bayesian Network (BN) [10], are widely used for image analysis. In addition, with improvements in spatial resolution, several works have considered spatial features, e.g., morphological profiles, attribute profiles [11][12][13]. Recently, the Bag-of-Word (BoW) model has been introduced for SAR image classification [14,15], which is inspired by the texton representation of an image, and this approach is based on the discriminative low-level features.…”