Genie Pro: robust image classification using shape, texture, and spectral information

Abstract

We present Genie Pro, a new software tool for image analysis produced by the ISIS (Intelligent Search in Images and Signals) group at Los Alamos National Laboratory. Like the earlier GENIE tool produced by the same group, Genie Pro is a general purpose adaptive tool that derives automatic pixel classification algorithms for satellite/aerial imagery, from training input provided by a human expert. Genie Pro is a complete rewrite of our earlier work that incorporates many new ideas and concepts. In particular, the new software integrates spectral information; and spatial cues such as texture, local morphology and large-scale shape information; in a much more sophisticated way. In addition, attention has been paid to how the human expert interacts with the software: Genie Pro facilitates highly efficient training through an interactive and iterative "training dialog". Finally, the new software runs on both Linux and Windows platforms, increasing its versatility. We give detailed descriptions of the new techniques and ideas in Genie Pro, and summarize the results of a recent evaluation of the software.

Automatic slag characterization based on hyperspectral image processing

“…Recent computer vision algorithms have proved that this kind of images is useful for image classification tasks [1] [3] [5] and automation tasks such as automatic metal garbage classification [2].…”

Assessing vegetation cover on the date of satellite-derived start of spring

“…The KAP imagery was converted to areal measures of vegetation cover fraction (VCF) using the evolving feature extraction algorithm of GENIE Pro software (Chantilly, VA, USA) (Perkins et al 2005). Each pixel was classified as either corn, bare soil or shadow and the percentage of each image occupied by corn was calculated (Viña et al 2004).…”

Fuzzy Spectral and Spatial Feature Integration for Classification of Nonferrous Materials in Hyperspectral Data

“…One problem that has to be addressed is the high dimensionality of the hyperspectral data [20]- [22]. In this sense, to circumvent the problems caused by the large resolution of the hyperspectral data when used for classification purposes, feature reduction techniques are usually applied to avoid the Hughes phenomenon [23].…”

“…In this sense, to circumvent the problems caused by the large resolution of the hyperspectral data when used for classification purposes, feature reduction techniques are usually applied to avoid the Hughes phenomenon [23]. To this end, the information contained in the spectral bands can be either decorrelated using principal component analysis (PCA) [20], wavelet decomposition [24]- [26], or by applying user-defined band selection [22], [27], [28]. Once the hyperspectral data is decorrelated, the feature vectors describing the spectrum are extracted and used for classification tasks.…”