Biologically-inspired data decorrelation for hyper-spectral imaging

Hyper-spectral data allows the construction of more robust statistical models to sample the material properties than the standard tri-chromatic color representation. However, because of the large dimensionality and complexity of the hyper-spectral data, the extraction of robust features (image descriptors) is not a trivial issue. Thus, to facilitate efficient feature extraction, decorrelation techniques are commonly applied to reduce the dimensionality of the hyper-spectral data with the aim of generating compact and highly discriminative image descriptors. Current methodologies for data decorrelation such as principal component analysis (PCA), linear discriminant analysis (LDA), wavelet decomposition (WD), or band selection methods require complex and subjective training procedures and in addition the compressed spectral information is not directly related to the physical (spectral) characteristics associated with the analyzed materials. The major objective of this article is to introduce and evaluate a new data decorrelation methodology using an approach that closely emulates the human vision. The proposed data decorrelation scheme has been employed to optimally minimize the amount of redundant information contained in the highly correlated hyper-spectral bands and has been comprehensively evaluated in the context of non-ferrous material classification

Biotic vs. Abiotic Control of Decomposition: A Comparison of the Effects of Simulated Extinctions and Changes in Temperature

The loss of species is known to have significant effects on ecosystem functioning, but only recently has it been recognized that species loss might rival the effects of other forms of environmental change on ecosystem processes. There is a need for experimental studies that explicitly manipulate species richness and environmental factors concurrently to determine their relative impacts on key ecosystem processes such as plant litter decomposition. It is crucial to understand what factors affect the rate of plant litter decomposition and the relative magnitude of such effects because the rate at which plant litter is lost and transformed to other forms of organic and inorganic carbon determines the capacity for carbon storage in ecosystems and the rate at which greenhouse gasses such as carbon dioxide are outgassed. Here we compared how an increase in water temperature of 5 degrees C and loss of detritivorous invertebrate and plant litter species affect decomposition rates in a laboratory experiment simulating stream conditions. Like some prior studies, we found that species identity, rather than species richness per se, is a key driver of decomposition, but additionally we showed that the loss of particular species can equal or exceed temperature change in its impact on decomposition. Our results indicate that the loss of particular species can be as important a driver of decomposition as substantial temperature change, but also that predicting the relative consequences of species loss and other forms of environmental change on decomposition requires knowledge of assemblages and their constituent species' ecology and ecophysiology.