Assessment Protocols and Comparison of Ordering Relations for Spectral Image Processing

Abstract

Recent developments in hyperspectral sensors allow to obtain high spectral and spatial resolutions that are close to the optical and physical structures of acquired surfaces. Consequently, hyperspectral imaging is used for its potential gain of accuracy. To preserve this metrological potential, generated bias, errors, and uncertainties must be managed at all subsequent processing levels. Based on the argument that a spectral image processing should avoid the linear approach, this study proposes several protocols for assessing the quality of spectral ordering relations. The protocols include considerations of theoretical properties to ensure result stability, in physical aspects of spectral processing to ensure the link to physical properties of materials, and experimental results using spectral images with physical ground truth to assess bias and errors. Full-band ordering relations are compared in order to find which satisfies all of the expected properties of a metrological spectral image processing.

Morphological Texture Description For Hyperspectral Images: Pattern Spectrum

“…This is because, at the center of its processing, MM requires determining the minimum and maximum values of a set of pixel values contained by the structuring element. In this work, we will use the conditional ratio and angular (CRA) ordering relation, which has been shown to be the most suitable for hyperspectral image processing in a previous study, respecting, in addition, the expected metrological properties [7]. CRA, as given in (3), is an ordering relation developed based on the ratio of distances relative to two spectral references S −∞ and S +∞ .…”

Full-Vector Gradient for Multi-Spectral or Multivariate Images

“…In addition, the extension to a multivariate case depends only on a color or spectral ordering [16]. Nevertheless, ordering often implies to select spectral or color references that impact the final result [17]- [19].…”

Gradient in spectral and color images: from the Di Zenzo initial construction to a generic proposition

“…An approach based on median filtering was proposed in the color domain from Astola et al [28]. To solve the difficulty of order n-dimensional vectors, numerous authors [31,32,40] presented the necessity to choose references. Selecting automatically the right reference is always an open question.…”