Marc Lennon

Support vector machines for hyperspectral image classification with spectral-based kernels

Support vector machines (SVM) has been recently used with success for the classification of hyperspectral images. This method appears to be a robust alternative for pattern recognition with hyperspectral data: since the method is based on a geometric point of view, no statistical estimation has to be achieved. Then, SVM outperforms classical supervised classification algorithms such as the maximum likelihood when the number of spectral bands increases or when the number of training samples remains limited. Nevertheless, those kernel-based methods do not take into consideration the spectral similarity between support vectors. Then, some modified kernels are presented to take into consideration the spectral similarity between support vectors to outperform SVM-based classification of hyperspectral data cube. Those kernels (that still suit Mercers conditions) are based on the use of spectral angle to evaluate the distance between support vectors. Classifiers to compare have been applied to an image from the CASI sensor including 17 bands from 450 to 950nm representing an intensive agricultural region (Brittany, France). It appears that those kernels reduce false alarms that were induced by illumination effects with classical kernels.

Hyperspectral image segmentation with Markov chain model

The Hidden Markov Chain (HMC) model has been extended to take into consideration the multi-component representation of an hyperspectral data cube. Parameters estimation is performed using the general Iterative Conditional Estimation (ICE) method. The vectorial extension of the model is straightforward since the vectorial point of view joints the observation of each pixel as a spectral signature. Then, the segmentation procedure achieves an estimation of multi-dimensional correlated probability density functions (pdf). Multi-dimensional densities have been estimated by a set of 1D densities through a projection step that makes component independent and of reduced dimension. Classifications have been applied to an image from the CASI sensor including 17 bands (from 450 to 950 nm) representing an intensive agricultural region (Brittany, France). Since, the intrinsic dimensionality of the observation has been estimated to 4, the multi-component HMC model has been applied to the CASI image reduced to 4 bands through an adapted projection pursuit method.

Oil slick detection and characterization by satellite and airborne sensors: experimental results with SAR, hyperspectral and lidar data

Efficient observation means are required for regional-scale detection of oil slicks at sea, as well as for local-scale quantitative mapping in order to support operational fight and recovering operations, including reliable choice and guidance of maritime and airborne fighting means. An efficient oil slick detection algorithm based on a multiscale approach is proposed for operational regional-scale detection from satellite SAR images. The potential of combining airborne passive hyperspectral imagery and active fluorescence laser technology is proposed for local-scale quantitative characterization. The ways towards the use of both satellites and airborne remote sensors for use in operational emergency scenarios are discussed.

Analysis of the reflectance spectra of oil emulsion spilled on the sea surface

Airborne remote sensing appears useful for monitoring oil spill accident or detecting illegal oil discharges. In that context, hyperspectral imagery in the SWIR range shows a high potential to describe oil spills. Indeed reflectance spectra of an oil emulsion layer show a wide variety of shapes according to its thickness or emulsion rate. Although based on laboratory measurements, it seems that these two parameters are insufficient to completely describe them. It appears that the way emulsion is performed leads to different reflectance spectra. Hence this paper will present a model which tends to simulate reflectance spectra of an oil emulsion layer over the sea water. To derive an analytical expression, some approximations and assumptions will be done. The result of this model shows high similarities with laboratory measurements and seems able to simulate most of the shapes of reflectance spectra. It also shows that a key parameter to define the shape of the reflectance spectra is the statistical distribution of water bubbles size in the emulsion. The description of this distribution function, if measurable, should be integrated into the methodology of elaboration of spectral libraries in the future.

Target Detection Using HF Radar Data

High frequency (HF) radar systems have a potential to detect targets which are located beyond the optical horizon on the sea surface. Unfortunately sea clutter may be strong enough to prevent such detection. In this work we present results of applying space-time adaptive processing (STAP) to this problem. Real, registered signals were used for numerical experiments.

Estimation of the sea bottom spectral reflectance in shallow water with hyperspectral data

The inversion of the semi-analytical radiative tranfer model proposed by Lee allows to estimate the water column paramters and depth. The solution generally used to deal with the sea bottom reflectance at the inversion stage is to use a spectral library containing relevant reflectance spectra of the studied area. In this paper, we will evaluate the influence of that spectral library. For this purpose, the simplest possible spectral library will be used: a spectral library including only one flat spectrum. The observations made using that spectral library will show results that re-evaluate the purpose of the spectral library itself at the inversion stage, leading us to propose a method to estimate the sea bottom spectral reflectance.