An experiment-based quantitative and comparative analysis of target detection and image classification algorithms for hyperspectral imagery

Over the past years, many algorithms have been developed for multispectral and hyperspectral image classification. A general approach to mi red pixel classification is linear spectral unmixing, which uses a linear mixture mod el to estimate the abundance fractions of signatures within a mixed pixel, As a result, the images generated for classification are usually gray scale images, where the gray level value of a pixel represents a combined amount of the abundance of spectral signatures residing in this pixel, Due to a l ack of standardized data, these mixed pixel algorithms have not been rigoro usly compared using a unified framework, In this paper, we present a compar ative study of some popular classification algorithms through a standardize d HYDICE data set with a custom-designed detection and classification crite rion. The algorithms to he considered for this study are those developed fo r spectral unmixing, the orthogonal subspace projection (OSP), maximum like lihood, minimum distance, and Fisher's linear discriminant analysis (LDA), In order to compare mixed pixel classification algorithms against pure pixe l classification algorithms, the mixed pixels are converted to pure ones by a designed mixed-to pure pixel converter, The standardized HYDICE data are then used to evaluate the performance of various pure and mixed pixel clas sification algorithms. Since all targets in the HYDICE image scenes can be spatially located to pixel level, the experimental results can be presented by tallies of the number of targets detected and classified for quantitati ve analysis.