A Proposal for the Management of the Measurement Uncertainty in Classification and Recognition Problems

Betta, G.; Capriglione, Domenico; Corvino, Mariella; Liguori, Consolatina; Paolillo, Alfredo

doi:10.1109/tim.2014.2347218

Cited by 10 publications

(4 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In more detail, the decision algorithm (see Figure 6) is composed by three steps [23], [25]: -STEP A: for each subject, j , present in the database, the measured Score, S j , is used together with the corresponding uncertainty, u S , in order to evaluate the probability, P j , that the input subject is the subject j. In particular, P j represents the probability that the Score of the j-th subject is equal to zero given a measured value (S j ), considering the score as a random variable (see Figure 6), whose standard deviation is equal to u S [25]; b) Figure 5. Definition of Wk: a) Images acquired by Camera 1 for subject #1, b) Scores evolution (the first image of the subject is used as reference).…”

Section: Classification Proceduresmentioning

confidence: 99%

“…The experimental activity demonstrated that the performance of these kinds of systems depends on several aspects from image acquisition to the classification procedure through the biometrical algorithm. The quantification of these uncertainty contributions lead the authors to formulate an original classification method which was able to improve the classification performance with respect to traditional scorebased approaches [25], [26].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Metrological characterization of 3D biometric face recognition systems in actual operating conditions

et al. 2017

Self Cite

View full text Add to dashboard Cite

<p>Nowadays, face recognition systems are going to widespread in many fields of application, from automatic user login for financial activities and access to restricted areas, to surveillance for improving security in airports and railway stations, to cite a few.<br />In such scenarios, the architectures based on stereo vision and 3D reconstruction of the face are going to assume a predominant role because they can generally assure a better reliability than solutions based on a single camera (which make use of a single image instead of a couple of images). To realize such systems, different architectures can be considered by varying the positioning of the pair of cameras with respect to the face of the subject to be identified, as well as both kind and resolution of camera considered. These parameters can affect the correct decision rate of the system in classifying the input face, especially in presence of image uncertainty.<br />In this paper, several 3D architectures differing in camera specifications and geometrical positioning of the camera pair (with respect to the input face) are realized and compared. The detection of facial features in the images is made by adopting a popular method based on the Active Appearance Model (AAM) algorithm. 3D position of facial features is then obtained by means of stereo triangulation. The performance of the realized systems has been compared in terms of sensitivity to the quantities of influence and related uncertainty, and of typical indexes for the analysis of classification systems. Main results of such comparison show that the best performance can be reached by reducing the distance between cameras and subject to be identified and by minimizing the horizontal angle between the plane containing the camera pair axis and the face to be identified.</p>

show abstract

Section: Classification Proceduresmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Metrological characterization of 3D biometric face recognition systems in actual operating conditions

et al. 2017

Self Cite

View full text Add to dashboard Cite

show abstract

“…With the aim of managing the measurement uncertainty for obtain a more reliable classification a new scheme was proposed in [13], [14] (see Figure 2): -At first, the two 2D images of the unknown subject are processed by the biometric algorithm and the Score, S i (i=1,..,N), of each of the N subjects of the database is evaluated. -…”

Section: The Classification Proceduresmentioning

confidence: 99%

“…The proposed method is based on a probabilistic approach for the evaluation of the confidence level of system outputs and the suitable use of this information for a performance improvement in terms of correct decision rate. In this way, the proposed method allows achieving a more reliable classification procedure with respect to a basic approach [14]. As cases study, it was applied to a 2D face classification algorithm based on Linear Discrimination Analysis.…”

Section: Introductionmentioning

confidence: 99%

A proposal for improving the performance of face recognition systems based on 3d features

Betta

Capriglione

Corvino

et al. 2015

2015 XVIII AISEM Annual Conference

View full text Add to dashboard Cite

In this paper a suitable methodology for the improvement of the reliability of results in classification systems based on 3D images is proposed. More in detail, it is based on the knowledge of the uncertainty of the features constituting the 3D image (obtained processing a pair of two 2D stereoscopic images) and on a suitable statistical approach providing a confidence level to the classification result. These pieces of information are then managed in order to improve the classification performance in terms of correct classification and missed classification percentages. The experimental results, obtained applying the methodology on an Active Appearance Models algorithm, a popular method for face recognition based on 3D features, show that, compared with a traditional approach (which generally does not take into account the uncertainty on 3D features), the proposed methodology allows to significantly improve the classification performance even in scenarios characterized by a high uncertainty.

show abstract

Performance Evaluation of Face Classification Systems

Betta

Capriglione

Corvino

et al. 2015

Lecture Notes in Electrical Engineering

Self Cite

View full text Add to dashboard Cite

In this paper, we propose a novel multiple path-metrics based multi-path (MPMP) routing protocol, which can find the most suitable route that satisfies the requirements of various data traffic in mobile ad hoc networks (MANETs). In order to provide differentiated QoS to various traffic, we propose a weighted cost function based on multiple path metrics, and D-ETX (Data-driven ETX) and available bandwidth estimation methods. To our best knowledge, this work is the first attempt to consider the multiple metrics-based multi-path selections for providing differentiated QoS to various traffic in MANETs. Simulation results demonstrate that the proposed protocol can achieve a significant improvement in terms of packet delivery ratio and end-to-end delay over existing protocols.

show abstract

A Proposal for the Management of the Measurement Uncertainty in Classification and Recognition Problems

Cited by 10 publications

References 23 publications

Metrological characterization of 3D biometric face recognition systems in actual operating conditions

Metrological characterization of 3D biometric face recognition systems in actual operating conditions

A proposal for improving the performance of face recognition systems based on 3d features

Performance Evaluation of Face Classification Systems

Contact Info

Product

Resources

About