Maciel Zortea

Recent Developments in Endmember Extraction and Spectral Unmixing

Spectral unmixing is an important task for remotely sensed hyperspectral data exploitation. The spectral signatures collected in natural environments are invariably a mixture of the pure signatures of the various materials found within the spatial extent of the ground instantaneous field view of the imaging instrument. Spectral unmixing aims at inferring such pure spectral signatures, called endmembers, and the material fractions, called fractional abundances, at each pixel of the scene. In this chapter, we provide an overview of existing techniques for spectral unmixing and endmember extraction, with particular attention paid to recent advances in the field such as the incorporation of spatial information into the endmember searching process, or the use of nonlinear mixture models for fractional abundance characterization. In order to substantiate the methods presented throughout the chapter, highly representative hyperspectral scenes obtained by different imaging spectrometers are used to provide a quantitative and comparative algorithm assessment. To address the computational requirements introduced by hyperspectral imaging algorithms, the chapter also includes a parallel processing example in which the performance of a spectral unmixing chain (made up of spatial–spectral endmember extraction followed by linear spectral unmixing) is accelerated by taking advantage of a low-cost commodity graphics co-processor (GPU). Combined, these parts are intended to provide a snapshot of recent developments in endmember extraction and spectral unmixing, and also to offer a thoughtful perspective on future potentials and emerging challenges in designing and implementing efficient hyperspectral imaging algorithms.

Automatic segmentation of dermoscopic images by iterative classification

Accurate detection of the borders of skin lesions is a vital first step for computer aided diagnostic systems. This paper presents a novel automatic approach to segmentation of skin lesions that is particularly suitable for analysis of dermoscopic images. Assumptions about the image acquisition, in particular, the approximate location and color, are used to derive an automatic rule to select small seed regions, likely to correspond to samples of skin and the lesion of interest. The seed regions are used as initial training samples, and the lesion segmentation problem is treated as binary classification problem. An iterative hybrid classification strategy, based on a weighted combination of estimated posteriors of a linear and quadratic classifier, is used to update both the automatically selected training samples and the segmentation, increasing reliability and final accuracy, especially for those challenging images, where the contrast between the background skin and lesion is low.

A computer aided diagnostic system for malignant melanomas

We describe a system for automatic diagnosis of malignant melanoma based on digital dermoscopic images. The tool is designed for use with general practitioners, saving time and resources in the diagnostic process. A variety of indicative features are described mimicking the human approach for diagnosis. Segmentation, pattern recognition, and change detection are the important steps in our approach.

Automatic learning of spatial patterns for diagnosis of skin lesions

We present a technique for automatic diagnosis of malignant melanoma based exclusively on local pattern analysis. The technique relies on local binary patterns in small sections in the image, and automatically selects the relevant texture features from those that discriminate best between benign and malignant skin lesions. The classification is performed using support vector machines, and the feature vectors are clustered using K-means clustering. The effects of K and window size are investigated. Reported average specificity and sensitivity are 73% for optimal parameter choice, indicating that the procedure is a useful part of a diagnostic system.

On the incorporation of spatial information to endmember extraction: Survey and algorithm comparison

Several well-known algorithms have been used for endmember extraction and spectral unmixing of hyperspectral imagery by considering only the spectral properties of the data when conducting the search. However, it might be beneficial to incorporate the spatial arrangement of the data in the development of endmember extraction and spectral unmixing algorithms. In this paper, we provide a survey on the use of spatial information in endmember extraction and further compare six different algorithms (three of which only use spectral information) in order to substantiate the impact of using spatial-spectral information versus only spectral information when searching for image endmembers. The comparison is carried out using a synthetic hyperspectral scene with spatial patterns generated using fractals, and a real hyperspectral scene collected by NASAs Airborne Visible Infra-Red Imaging Spectrometer (AVIRIS).

Experiments on feature extraction in remotely sensed hyperspectral image data

In the present study, we propose a new simple approach to reduce the data dimensionality in hyperspectral image data. The basic assumption here consists in assuming that a pixels curve of spectral response, as defined in the spectral space by the recorded digital numbers (DNs) at the available spectral bands, can be segmented, and each segment can be replaced by a smaller number of statistics: mean and variance, describing the main characteristics of a pixels spectral response. It is expected that this procedure can be accomplished without significant loss of information. The DNs at even spectral band are used to calculate a few statistics that would be used instead of the DNs themselves in the classification process. For the pixels spectral curve segmentation, we propose tree sub-optimal algorithms that are easy to implement and also computationally efficient. Using a top-down strategy, the original pixels spectral curve is sequentially segmented. Experiments using a parametric classifier are performed on an AVIRIS data set. Encouraging results have been obtained in terms of classification accuracy and execution time, suggesting the effectiveness of the proposed algorithms. The results suggest that the proposed algorithms can be faster and achieve a better accuracy than the classical Sequential Forward Selection (SFS) technique, known from literature as one of the simplest and fastest techniques for data dimensionality reduction using the feature selection approach.

Computer-Aided Decision Support for Melanoma Detection Applied on Melanocytic and Nonmelanocytic Skin Lesions: A Comparison of Two Systems Based on Automatic Analysis of Dermoscopic Images

Commercially available clinical decision support systems (CDSSs) for skin cancer have been designed for the detection of melanoma only. Correct use of the systems requires expert knowledge, hampering their utility for nonexperts. Furthermore, there are no systems to detect other common skin cancer types, that is, nonmelanoma skin cancer (NMSC). As early diagnosis of skin cancer is essential, there is a need for a CDSS that is applicable to all types of skin lesions and is suitable for nonexperts. Nevus Doctor (ND) is a CDSS being developed by the authors. We here investigate NDs ability to detect both melanoma and NMSC and the opportunities for improvement. An independent test set of dermoscopic images of 870 skin lesions, including 44 melanomas and 101 NMSCs, were analysed by ND. Its sensitivity to melanoma and NMSC was compared to that of Mole Expert (ME), a commercially available CDSS, using the same set of lesions. ND and ME had similar sensitivity to melanoma. For ND at 95% melanoma sensitivity, the NMSC sensitivity was 100%, and the specificity was 12%. The melanomas misclassified by ND at 95% sensitivity were correctly classified by ME, and vice versa. ND is able to detect NMSC without sacrificing melanoma sensitivity.

Automatic tree crown delineation in tropical forest using hyperspectral data

This paper aims to use unique features of hyperspectral data on an automatic process for outlining individual tree crowns (ITCs) in a tropical forest area, with special focus on semi-deciduous species. In order to enhance biophysical and biochemical properties of canopy species, a set of vegetation indices were computed. These indices served as input for a region growing segmentation algorithm that takes into account mutual similarity of pixels and spectral separability between neighbor segments. Segmentation output was evaluated on the basis of a score computed with the proportion of the area of the segments located within manually delineated ITCs. Results show that the segmentation approach is able to automatically delineate up to 70% of the control ITCs.

Spectral-textural endmember extraction

Several available techniques for endmember extraction and spectral unmixing use only the spectral information contained in the hyperspectral data. In this paper, we introduce a novel method for spatial-spectral endmember extraction which incorporates texture features in the quantification of spatial information (jointly with spectral information). Experimental results with simulated and real hyperspectraldata sets indicate that textural information could assist the extraction of spectral endmembers, although a challenging issue still remains: how to combine the final set of endmember candidates (obtained by merging the individual sets of candidates found using spectral, textural and joint spectral-textural information) in order to provide a relevant final solution.

Spatial-spectral endmember extraction from remotely sensed hyperspectral images using the watershed transformation

In this paper, we investigate the use of the watershed transformation for integrating spatial and spectral information in the process of endmember extraction for spectral unmixing of hyperspectral images. The proposed approach is presented as a preprocessing module designed to automatically select a small subset of pixels containing potentially relevant candidates from both spatial and spectral point of view. Dimensionality reduction is required. The idea is to use the morphological watershed transformation to guide the endmember searching process to spatially homogeneous and spectrally “purer” areas. Here the main assumption is that such areas can be located at the local minima of the catchment basins, and far away from watershed lines that define the transition areas between different regions, expected to contain mixed pixels. Experimental results, conducted using a database of 28 simulated hyperspectral data sets obtained through manipulation of a real hyperspectral image acquired by the Airborne Visible Infra-Red Imaging Spectrometer (AVIRIS) over a mixed scenario including agricultural, vegetation, and urban areas, suggests a promising trade-off between percentage of endmember candidates retained and degree of spectral purity of predominant endmembers.