Narrow Results

The biometric verification task is to determine whether or not an input and a template belong to the same individual. In the context of automatic fingerprint verification the task consists of three steps: feature extraction, where features (typically minutiae) are extracted from each fingerprint, scoring, where the degree of match between the two sets of features is determined, and decision, where the score is used to accept or reject the hypothesis that the input and template belong to the same individual. The paper focuses on the final decision step, which is a binary classification problem involving a single score variable. The commonly used decision method is to learn a score threshold from a labeled set of inputs and templates, by first determining the receiver operating characteristic (ROC) of the task. The ROC method works well when there is a well-registered fingerprint image. The paper shows that when there is uncertainty due to fingerprint quality, e.g. the input is a latent print or a partial print, the decision method can be improved by using the likelihood ratio of match/non match. The likelihood ratio is obtained by modeling the distributions of same finger and different finger scores using parametric distributions. The parametric forms considered are Gaussian and Gamma distributions whose parameters are learnt from labeled training samples. The performances of the likelihood and ROC methods are compared for varying numbers of minutiae points available for verification. Using either Gaussian or Gamma parametric distributions, the likelihood method has a lower error rate than the ROC method when few minutiae points are available. Likelihood and ROC methods converge to the same accuracy as more minutiae points are available.

The paper describes an integrated recognition-by-parts architecture for reliable and robust face recognition. Reliability and robustness are characteristic of the ability to deploy full-fledged and operational biometric engines, and handling adverse image conditions that include among others uncooperative subjects, occlusion, and temporal variability, respectively. The architecture proposed is model-free and non-parametric. The conceptual framework draws support from discriminative methods using likelihood ratios. At the conceptual level it links forensics and biometrics, while at the implementation level it links the Bayesian framework and statistical learning theory (SLT). Layered categorization starts with face detection using implicit rather than explicit segmentation. It proceeds with face authentication that involves feature selection of local patch instances including dimensionality reduction, exemplar-based clustering of patches into parts, and data fusion for matching using boosting driven by parts that play the role of weak-learners. Face authentication shares the same implementation with face detection. The implementation, driven by transduction, employs proximity and typicality (ranking) realized using strangeness and p-values, respectively. The feasibility and reliability of the proposed architecture are illustrated using FRGC data. The paper concludes with suggestions for augmenting and enhancing the scope and utility of the proposed architecture.

We have investigated different scenarios of combining pattern matchers. The combination problem can be viewed as a construction of a postprocessing classifier operating on the matching scores of the combined matchers. The optimal combination algorithm for verification systems corresponds to the likelihood ratio combination function. It can be implemented by the direct reconstruction of this function with genuine and impostor score density approximations. However, the optimal combination algorithm for identification systems is difficult to express analytically. We will show that this difficulty is caused by the dependencies between matching scores assigned to different classes by the same classifier. The experiments on the large sets of scores from handwritten word recognizers operating on postal images and biometric matchers (NIST biometric score set BSSR1) confirm the existence of such dependencies and that the optimal combination functions for verification and identification systems are different.

In this paper, we introduce two new confidence measures for large vocabulary speech recognition systems. The major feature of these measures is that they can be computed without waiting for the end of the audio stream. We proposed two kinds of confidence measures: frame-synchronous and local. The frame-synchronous ones can be computed as soon as a frame is processed by the recognition engine and are based on a likelihood ratio. The local measures estimate a local posterior probability in the vicinity of the word to analyze. We evaluated our confidence measures within the framework of the automatic transcription of French broadcast news with the EER criterion. Our local measures achieved results very close to the best state-of-the-art measure (EER of 23% compared to 22.0%). We then conducted a preliminary experiment to assess the contribution of our confidence measure in improving the comprehension of an automatic transcription for the hearing impaired. We introduced several modalities to highlight words of low confidence in this transcription. We showed that these modalities used with our local confidence measure improved the comprehension of automatic transcription.

In firearm identification, a firearm examiner looks at a pair of fired bullets or cartridge cases using a comparison microscope and determines from this visual analysis if they were both fired from the same firearm. In the particular case of fired bullets, the individual firearm signature takes the form of a striated pattern. Over the time, the firearm examiner’s community developed two distinct approaches for bullet identification: pattern matching and line counting. More recently, the emergence of technology enabling the capture of surface topographies down to a submicron depth resolution has been a catalyst for the field of computerized objective ballistic identification. Objectiveness is achieved through the statistical analysis of various scores of known matches and known nonmatches exhibit pair comparison, which in turn implies the capture of large quantities of bullets and cartridge cases topographies. The main goal of this study was to develop an objective identification method for bullets fired from conventionally rifled barrels, and to test this method on public and proprietary bullet 3D image datasets captured at different lateral resolutions. Two newly developed bullet identification scores, the Line Counting Score (LCS) and the Pattern Matching Score, computed on 3D topographies yielded perfect match versus nonmatch separation for three different sets used in the standard Hamby–Brundage Test. A similar analysis performed using a larger, more-realistic set, enabled us to define a discriminative line at a false match rate of 1/10000 on a 2D plot that shows both identification scores for matches and nonmatches. The LCS is shown to produce a better sensitivity than the standard consecutive matching striae criteria for the more-realistic dataset. A likelihood function was also computed from a linear combination of both scores, and a conservative approach based on extreme value theory is proposed to extrapolate this function in the score domain where nonmatch data are not available. This study also provides a better understanding of the limitations of studies that involve very few firearms.

The occurrence of laboratory typing error in pedigree data collected for use in linkage analysis cannot be ignored. In maps where recombinations between nearby markers rarely occur, each erroneous recombinations (result of typing error) is given substantial weight thereby increasing the estimate of θ, the recombination fraction. As the maps being developed become more dense, θ approaches the error rate and most of all observed crossovers will be erroneous. We present a method for detecting errors in pedigree data. The index is a variant of the likelihood ratio test statistic and is used to test the null hypothesis of no error for an individual at a locus versus the alternative hypothesis of error. High values of the index correspond to unlikely genotypes. The method has been shown to detect errors introduced into CEPH pedigrees and an error in a larger experimental pedigree (retinitis pigmentosa). While the method was designed to detect typing error, it is sufficiently general to detect any relatively unlikely genotype and therefore can also be used to detect pedigree error.

Statistical process control charts are intended to assist operators in detecting process changes. If a process change does occur, the control chart should detect the change quickly. If the operator is provided with an estimate as to when the process changed, the search to find the special cause can be more easily facilitated. We investigate a process-monitoring tool for Poisson count data that quickly responds to process mean count rate changes regardless of the magnitude of the change, while supplying useful diagnostic information. A likelihood ratio approach was used to develop a control chart for a permanent step change in a Poisson process rate parameter. The average run length (ARL) performance of this chart is compared to that of several Poisson cumulative sum (CUSUM) control charts. Our performance results show that the proposed chart performs better than any one CUSUM chart over a wide range of potential shift magnitudes. The proposed chart also provides maximum likelihood estimates of the time and the magnitude of the process shift. These crucial change point diagnostics can greatly enhance the special cause investigation.

We measure the performance of probabilistic models from a decision-theoretic perspective along the lines of Friedman and Sandow [6]. In particular, we adopt the point of view of an investor who evaluates models based on the test-sample averaged utility of the expected-utility-optimal strategies that the models suggest in the horse race setting. In this paper, we relax the assumptions of Friedman and Sandow [6]: we omit the notion of a "true" measure and we allow our investor to withhold or borrow cash, which widens the range of possible applications. We show that, in this setting, our relative model performance measure is odds-ratio independent if and only if the investor has a generalized logarithmic utility function, in which case it essentially reduces to the likelihood ratio. We also show that for horse races with nearly homogeneous returns, our relative performance measure is approximately equal to the likelihood ratio.

The location of a passage is a kind of semantic information that may prove useful for a variety of applications dealing with inference over passages described in natural language texts. In this paper, we propose a method for automatic discovery of pairs of an event and a place term related by the location, such as wash clothes ⇒ laundry room. In contrast to previous approaches that extracting associations between particular actions and locations when those actions occur on dunning likelihood ratio, the underlying assumption of our method is that correlation between an event and a place term is in the regular co-occurrence of a verb and a place noun within locally coherent text. Based on the analogy with the problem of inferring semantic information from text corpus statistical method of Dunning's likelihood ratio is used to score the extracted pairs for association in order to discover the correlation in each pair. In our experimental evaluation, we examine the effect that various statistical methods produce on the accuracy of this model of inferring locations. After that we carried out a direct evaluation of rare pairs against different statistical method.

The Edgeworth expansion of the log-likelihood ratio by the maximum likelihood method in the Vasicek model is studied in this paper. Our method provides an improvement on the accuracy of reverting speed estimation under finite time data. The error bound is also provided. The Edgeworth expansion for the distribution is helping with this improvement. Hence, the reverting speed of interest rates can achieve a better simulation result based on such modifications.