Towards the definition of a flexible hyperspectral processing chain: Preliminary case study using high-resolution urban data

The Hyperspectral Imaging Network (Hyper-I-Net) is a four-year Marie Curie research training network designed to build an interdisciplinary European research community focused on hyperspectral imaging activities [1]. One of the main activities of Hyper-I-Net is to settle the basis for the definition and testing of a flexible hyperspectral data collection and processing chain, in which individual elements can be integrated in such a way that the resulting chain can be dynamically adapted and reconfigured to satisfy the requirements of different application scenarios with little effort [2]. Since efficient hyperspectral data processing can be a really complex procedure, we approach this problem in the context of a multidisciplinary collaboration so that the proposed activity can benefit from the complementary skills of partners with expertise in heterogeneous disciplines such as sensor design and calibration [3], pattern recognition, signal and image processing [4], and Earth observation related products [5]. The outcome of this joint activity is expected to be a set of hardware/software processing techniques able to deal with the intrinsic complexity of hyperspectral data in an effective manner. In this paper, we describe a first approximation to the relevant issue of defining a part of the hyperspectral processing chain in a flexible manner. An ultimate goal of our study is to objectively quantify the impact of different (standard and new) processing stages on the generation of a realistic, user-oriented product in the context of an urban land cover mapping problem by means of hyperspectral data, selected in this work as an application case study for demonstration purposes. Specifically, the processing steps considered in our framework include [6] dimensionality reduction, feature selection, feature extraction and classification. The impact of the processing modules above is objectively quantified in this work by implementing several processing chains made up of different combinations of such modules. Each chain is then thoroughly evaluated using hyperspectral data sets collected at multiple spatial and spectral resolutions by the DAIS 7915 and ROSIS imaging spectrometers over a well-known urban area in Pavia, Italy. Although the proposed study is linked to a specific application domain, our experimental results reveal interesting considerations that may help image analysts in defining customizedprocessing chains based on parameters which can be identified and objectively evaluated a priori, such as available sensor resolution or ancillary information. In addition, our study also demonstrates the importance of incorporating information related to both the spatial and the spectral domain in the different steps that comprise the hyperspectral processing chain; particularly when such chain can take advantage of the combined use of both sources of information as it is the case in the considered urban characterization application.