Search NASA⌕ Search

Engineering topics

Gupta, Shalini

Publications and source records attributed to Gupta, Shalini.

Reducing variability in the output of pattern classifiers using histogram shaping

Purpose: The authors present a novel technique based on histogram shaping to reduce the variability in the output and (sensitivity, specificity) pairs of pattern classifiers with identical ROC curves, but differently distributed outputs. Methods: The authors identify different sources of variability in the output of linear pattern classifiers with identical ROC curves, which also result in classifiers with differently distributed outputs. They theoretically develop a novel technique based on the matching of the histograms of these differently distributed pattern classifier outputs to reduce the variability in their (sensitivity, specificity) pairs at fixed decision thresholds, and to reduce the variability in their actual output values. They empirically demonstrate the efficacy of the proposed technique by means of analyses on the simulated data and real world mammography data. Results: For the simulated data, with three different known sources of variability, and for the real world mammography data with unknown sources of variability, the proposed classifier output calibration technique significantly reduced the variability in the classifiers' (sensitivity, specificity) pairs at fixed decision thresholds. Furthermore, for classifiers with monotonically or approximately monotonically related output variables, the histogram shaping technique also significantly reduced the variability in their actual output values. Conclusions: Classifier output calibration based on histogram shaping can be successfully employed to reduce the variability in the output values and (sensitivity, specificity) pairs of pattern classifiers with identical ROC curves, but differently distributed outputs.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Advanced Structural and Inflatable Hybrid Spacecraft Module

An inflatable module comprising a structural core and an inflatable shell, wherein the inflatable shell is sealingly attached to the structural core. In its launch configuration, the wall thickness of the inflatable shell is collapsed by vacuum. Also in this configuration, the inflatable shell is collapsed and efficiently folded around the structural core. Upon deployment, the wall thickness of the inflatable shell is inflated; whereby the inflatable shell itself, is thereby inflated around the structural core, defining therein a large enclosed volume. A plurality of removable shelves are arranged interior to the structural core in the launch configuration. The structural core also includes at least one longeron that, in conjunction with the shelves, primarily constitute the rigid, strong, and lightweight load-bearing structure of the module during launch. The removable shelves are detachable from their arrangement in the launch configuration so that, when the module is in its deployed configuration and launch loads no longer exist, the shelves can be rearranged to provide a module interior arrangement suitable for human habitation and work. In the preferred embodiment, to provide efficiency in structural load paths and attachments, the shape of the inflatable shell is a cylinder with semi-toroidal ends.

Schneider, William C.↗

ERBE Geographic Scene and Monthly Snow Data

The Earth Radiation Budget Experiment (ERBE) is a multisatellite system designed to measure the Earth's radiation budget. The ERBE data processing system consists of several software packages or sub-systems, each designed to perform a particular task. The primary task of the Inversion Subsystem is to reduce satellite altitude radiances to fluxes at the top of the Earth's atmosphere. To accomplish this, angular distribution models (ADM's) are required. These ADM's are a function of viewing and solar geometry and of the scene type as determined by the ERBE scene identification algorithm which is a part of the Inversion Subsystem. The Inversion Subsystem utilizes 12 scene types which are determined by the ERBE scene identification algorithm. The scene type is found by combining the most probable cloud cover, which is determined statistically by the scene identification algorithm, with the underlying geographic scene type. This Contractor Report describes how the geographic scene type is determined on a monthly basis.

Coleman, Lisa H.↗