Detection of gaseous plumes in IR hyperspectral images using hierarchical clustering

Abstract

The emergence of IR hyperspectral sensors in recent years enables their use in remote environmental monitoring of gaseous plumes. IR hyperspectral imaging combines the unique advantages of traditional remote sensing methods such as multispectral imagery and nonimaging Fourier transform infrared spectroscopy, while eliminating their drawbacks. The most significant improvement introduced by hyperspectral technology is the capability of standoff detection and discrimination of effluent gaseous plumes without need for a clear reference background or any other temporal information. We introduce a novel approach for detection and discrimination of gaseous plumes in IR hyperspectral imagery using a divisive hierarchical clustering algorithm. The utility of the suggested detection algorithm is demonstrated on IR hyperspectral images of the release of two atmospheric tracers. The application of the proposed detection method on the experimental data has yielded a correct identification of all the releases without any false alarms. These encouraging results show that the presented approach can be used as a basis for a complete identification algorithm for gaseous pollutants in IR hyperspectral imagery without the need for a clear background.

Determination of optical and microphysical properties of thin warm clouds using ground based hyper-spectral analysis

“…The presented approach treats thin warm clouds as semi transparent objects which alter the magnitude and the spectrum of the clear background signal. Therefore, we have used techniques borrowed from the field of remote detection and identification of gaseous and aerosols plumes for environmental applications, under the assumption that they are the most suitable tools to retrieve thin clouds' properties (see for example, Hirsch andAgassi, 2007, andAgassi et al, 2008). The analysis presented in this study relies on 3 elements: a radiation transfer model, a method to extract the effect of a water cloud while eliminating the sky background, and a spectral matching method.…”

“…Therefore, in order to enhance the cloud's signature, we subtract the clear sky signal from the obtained spectra. The method of clear reference subtraction is commonly applied for detection and identification of weak gaseous plumes (Hirsch and Agassi, 2007). It enables retrieval of only the differential spectral signal, and therefore analyzes the phenomenon which created the signal without considering the background.…”

A machine learning approach for material detection in hyperspectral images

“…Although invisible to the human eye, different materials (gases, liquids, solids) exhibit unique spectral signatures. Automatic detection of these traces is of great importance, for example for environmental monitoring where possible applications include the detection of oil slicks in the sea after a shipwreck [7] or the detection of industrial gaseous pollutants [5]. In the biomedical field, hyperspectral image analysis approaches could help to detect abnormalities such as skin tumours [2,6] or other malignancies in tissues [10].…”

Low-Rank Decomposition and Total Variation Regularization of Hyperspectral Video Sequences

“…But in HVS, these methods can only be applied frame by frame and the post processing technique might be different in different scenes. The second category performs clustering of spectral data to separate the plume from the background [8,[19][20]. These methods often exhibit more spatial continuity.…”

“…In an earlier paper [8], it was shown that principal component analysis applied to each frame resulting in temporal flicker for the video, requiring a Midway equalization procedure. The later works [19,20] are able to perform graphbased clustering across several video frames, without this effect. The third category is to extend the traditional object tracking algorithms to HVS.…”