RETRACTED: Application of Bayes linear discriminant functions in image classification

2012 ◽  
Vol 33 (3) ◽  
pp. 278-282 ◽  
Author(s):  
Kęstutis Dučinskas ◽  
Lijana Stabingienė ◽  
Giedrius Stabingis
Keyword(s):  
Image Classification ◽  
Discriminant Functions ◽  
Linear Discriminant

scholarly journals Comparison of linear discriminant functions in image classification

2010 ◽  
Vol 51 ◽  
Author(s):  
Lijana Stabingienė ◽  
Giedrius Stabingis ◽  
Kęstutis Dučinskas
Keyword(s):  
Random Field ◽  
Image Classification ◽  
Spatial Correlation ◽  
Conditional Distribution ◽  
Gaussian Random Field ◽  
Training Sample ◽  
Discriminant Functions ◽  
Linear Discriminant ◽  
And Performance ◽  
Bayes Discriminant Function

In statistical image classification it is usually assumed that feature observations given labels are independently distributed. We have retracted this assumption by proposing stationary Gaussian random field (GRF) model for features observations. Conditional distribution of label of observation to be classified is assumed to be dependent on its spatial adjacency with training sample spatial framework. Perfomance of the Bayes discriminant function (BDF) and performance of plug-in BDFare tested and are compared with ones ignoring spatial correlation among feature observations.For illustration image of figure corrupted by additive GRF is analyzed. Advantage of proposed BDF against competing ones is shown visually and numerically.

DECISION MAKING BY VARIABLE CONTRIBUTION IN DISCRIMINANT, LOGIT, AND REGRESSION ANALYSES

2004 ◽  
Vol 03 (02) ◽  
pp. 265-279 ◽  
Author(s):  
STAN LIPOVETSKY ◽  
MICHAEL CONKLIN
Keyword(s):  
Decision Making ◽  
Logistic Regression ◽  
Regression Analysis ◽  
Statistical Models ◽  
Regression Analyses ◽  
Discriminant Functions ◽  
Multivariate Statistical ◽  
Linear Discriminant ◽  
Variable Contribution ◽  
Linear Aggregate

Comparative contribution of predictors in multivariate statistical models is widely used for decision making on the importance of the variables for the aims of analysis and prediction. However, the analysis can be made difficult because of the predictors' multicollinearity that distorts estimates for coefficients in the linear aggregate. To solve the problem of the robust evaluation of the predictors' contribution, we apply the Shapley Value regression analysis that provides consistent results in the presence of multicollinearity both for regression and discriminant functions. We also show how the linear discriminant function can be constructed as a multiple regression, and how the logistic regression can be approximated by linear regression that helps to obtain the variables contribution in the linear aggregate.

Linear Discriminant Functions to Improve the Glaucoma Probability Score Analysis to Detect Glaucomatous Optic Nerve Heads

2013 ◽  
Vol 22 (2) ◽  
pp. 73-79 ◽  
Author(s):  
Michele Iester ◽  
Francesco Oddone ◽  
Mirko Prato ◽  
Marco Centofanti ◽  
Paolo Fogagnolo ◽  
...  
Keyword(s):  
Optic Nerve ◽  
Discriminant Functions ◽  
Linear Discriminant ◽  
Probability Score ◽  
Score Analysis ◽  
Glaucoma Probability Score

ODOR MIXTURES AND CHEMOSENSORY ADAPTATION IN GAS SENSOR ARRAYS

2003 ◽  
Vol 12 (01) ◽  
pp. 1-16 ◽  
Author(s):  
RICARDO GUTIERREZ-OSUNA ◽  
NILESH U. POWAR
Keyword(s):  
Chemical Sensor ◽  
Sensor Arrays ◽  
Ternary Mixtures ◽  
Discriminant Functions ◽  
Linear Discriminant ◽  
Chemical Sensor Array ◽  
Binary And Ternary Mixtures ◽  
Olfactory Systems ◽  
Odor Mixtures ◽  
Input Adaptation

Inspired by the process of olfactory adaptation in biological olfactory systems, this article presents two algorithms that allow a chemical sensor array to reduce its sensitivity to odors previously detected in the environment. The first algorithm is based on a committee machine of linear discriminant functions that operate on multiple subsets of the overall sensory input. Adaptation occurs by depressing the voting strength of discriminant functions that display higher sensitivity to previously detected odors. The second algorithm is based on a topology-preserving linear projection derived from Fisher's class separability criteria. In this case, the process of adaptation is implemented through a reformulation of the between-to-within-class scatter eigenvalue problem. The proposed algorithms are validated on two datasets of binary and ternary mixtures of organic solvents using an array of temperature-modulated metal-oxide chemoresistors.

The performance of delta check methods.

1979 ◽  
Vol 25 (12) ◽  
pp. 2034-2037 ◽  
Author(s):  
L B Sheiner ◽  
L A Wheeler ◽  
J K Moore
Keyword(s):  
False Positive ◽  
False Positive Rate ◽  
True Positive Rate ◽  
True Positive ◽  
Discriminant Functions ◽  
Linear Discriminant ◽  
Operating Rate ◽  
Positive Rate ◽  
The Relationship

Abstract The percentage of mislabeled specimens detected (true-positive rate) and the percentage of correctly labeled specimens misidentified (false-positive rate) were computed for three previously proposed delta check methods and two linear discriminant functions. The true-positive rate was computed from a set of pairs of specimens, each having one member replaced by a member from another pair chosen at random. The relationship between true-positive and false-positive rates was similar among the delta check methods tested, indicating equal performance for all of them over the range of false-positive rate of interest. At a practical false-positive operating rate of about 5%, delta check methods detect only about 50% of mislabeled specimens; even if the actual mislabeling rate is moderate (e.g., 1%), only abot 10% of specimens flagged a by a delta check will actually have been mislabeled.

Sign in / Sign up

Export Citation Format

Share Document