Elimination of information redundancy of hyperspectral images using the “well-adapted” basis method

Abstract

The paper considers some issues of eliminating information redundancy of hyperspectral images (HSI). The characteristic properties of the HSI are listed, a brief description of the existing HSI compression methods is given. The possibility of using local, homogeneous “well-adapted” basis functions (LHWABF) to eliminate information redundancy and adaptive compression of the HSI is considered. An algorithm for constructing a LHWABF system for the HSI based on the Chebyshev approximation is proposed. The results of computational experiments, including the use of a graphics processor, are presented. The effectiveness of the proposed method of adaptive compression HSI is shown.

Elimination of Information Redundancy of Hyperspectral Raster Images Technology Enhancement

“…In [12], the authors proposed a sequential quasi-optimal algorithm for eliminating the information redundancy of raster hyperspectral images of remote sensing of the Earth with controlled losses based on the construction of a system of local homogeneous "well-adapted" basis functions [17,18]. Practical use of the algorithm revealed its sufficient efficiency in terms of the obtained compression ratio of raster hyperspectral images, however, the temporal efficiency of its operation can hardly be called satisfactory.…”

“…It was shown in [12] that the process of constructing the unit vectors of a system of local homogeneous "well-adapted" basis functions consists of a set of stages described by formulas (1) and (2):…”

Description Models, Methods, Algorithms, and Technology for Processing Poorly tructured Raster Graphic Documents

“…This method provides an encoding error that is not greater than the one specified for all points of the convex hull of the original set of vectors, while methods based on the ideology of the main components provide a sufficiently small error on average for the entire source set. The proposed method allows for a fairly simple practical implementation for large dimensions of the source data [28].…”