Showing papers on "Mahalanobis distance published in 1977"
TL;DR: In some ranking and selection problems, it is reasonable to consider any population which is inferior but sufficiently close to the best (t-th best) as acceptable as acceptable as discussed by the authors.
Abstract: In some ranking and selection problems it is reasonable to consider any population which is inferior but sufficiently close to the best (t-th best) as acceptable. Under this assumption, this paper studies classes of procedures to meet two possible goals. A and B. Goal A is to select a subset which contains only good populations, while Goal B is of a screening nature and requires selection of a subset of size not exceeding m (1 ≤ m ≤ k) and containing at least one good population. In each case results loading to the determination of the sample size required to attain the goals above with prespecified probability are obtained. Properties of the procedures are discussed.
13 citations
TL;DR: The lower bound on the probability of correct classification is a monotonically increasing function of the Mahalanobis distance for all monotonic ellipsoidally symmetric forms.
8 citations
01 May 1977
TL;DR: A number of distance measures for speaker independent recognition of isolated words are proposed, which use auto-correlation coefficients alone or autocorrelation and linear predictor coefficients as feature paramaters of the speech samples.
Abstract: In this paper a number of distance measures for speaker independent recognition of isolated words are proposed. These distance measures use auto-correlation coefficients alone or autocorrelation and linear predictor coefficients as feature paramaters of the speech samples. One measure used is the measure in discrete l 2 space of linear functionals. Some of the other distance measures used are the "nearest in angle" or the normalized correlation measure and the Mahalanobis distance. Actual evaluation of these distance measures is then performed using a standard 40 word reading test vocabulary spoken by 25 different speakers. All the above measures give good recognition results. The best distance measure has given a recognition rate of 87.3%.
2 citations
01 Aug 1977
TL;DR: In this article, a new texture measurement and the Mahalanobis classifier with a generalized inverse scheme were developed to generate decision maps of terrain features with digitized B/W photographs on a pixel by pixel basis.
Abstract: : A new texture measurement and the Mahalanobis classifier with a generalized inverse scheme were developed to generate decision maps of terrain features with digitized B/W photographs on a pixel by pixel basis. Eight scences within the Northeast test area, four low altitude and four high altitude, were analyzed yielding a hit-rate of about 90% with properly digitized image data. To determine the degree of non-normal behavior of the texture variables, the stable distribution models were utilized. Methods of estimating the stable parameters of the texture variables were developed. It is found that fifty % of the texture variables are not normally distributed. Since the stable distribution models are capable of incorporating the skewness parameters into the classification process, it is recommended as a new classifier for image data analysis.
1 citations
01 Dec 1977
TL;DR: A nonparametric maneuver detector is implemented in a Kalman filter to enhance the capability of tracking maneuvering targets by comparison with a standard maneuver detector which uses the Mahalanobis distance.
Abstract: A nonparametric maneuver detector is implemented in a Kalman filter to enhance the capability of tracking maneuvering targets. Maneuver detection is done independently in both x- and y-coordinates and appropriate corrections are made to the error covariance matrix. A comparison is made between the nonparametric maneuver detector and a standard maneuver detector which uses the Mahalanobis distance.