The Gabor transform basis of chromatic monitoring

Abstract

Chromaticity based monitoring methodologies have evolved from polychromatic optical fibre based sensing systems, which was first implemented during the 1980s. The evolution has led to a recognition that such a methodology is applicable to monitoring systems more widely. It is shown that the method may be regarded as the manifestation of a particular form of signal representation by Gabor transforms. Examples of generically different practical implementations of the chromatic methodology to a range of different monitoring situations are given.

Chromatic Monitoring of Gear Mechanical Degradation Based on Acoustic Emission

“…The chromatic analysis is based on the interpretation of applying different receptors partially superposed over a signal [17], [18], like human perception. The resulting parameters are, then, interconnected between them, which allow a smooth behavior of their response.…”

Investigation of Relations Between Skin Cancer Lesions‘ Images and Their Reflectance and Fluorescent Spectra

“…Consequently, the spectrum of white light passing through such a tissue can be affected by the various components and thus become biased towards the longer wavelengths (Jones, 2000). To benefit fully from optical spectroscopy's advantages, one needs to relate the spectral features with the morphology and biochemical composition of the tissue investigated.…”

Invariant spatial chromatic processors for region image description

“…The outputs from these processors are used to form cross correlations between the signal and the different processors responses [7]. Normally each processor response may have a Gaussian shape aligning with the Gabor transform [8], a triangle or other shapes may be preferred for some cases when a special type of responsivity is required. Whatever the processor shape, the non-orthogonality is significant in cross correlating the subdivided signal on each channel and in providing a means of selective tuning in the regions of overlap.…”

“…Whatever the processor shape, the non-orthogonality is significant in cross correlating the subdivided signal on each channel and in providing a means of selective tuning in the regions of overlap. It has been shown that although three spatial chromatic processors provide a high level of signal discrimination, additional processors could enhance signal discrimination even further but that extending the number beyond six gives little advantage [8], [9]. For this reason and for computing simplicity three half height overlapping Gaussian processors 'HHOGPs' are used in this paper.…”