Cesar San Martin

Face Recognition with LWIR Imagery Using Local Binary Patterns

In this paper, the merits of the Local Binary Patterns (LBP) representation are investigated in the context of face recognition using long-wave infrared images. Long-wave infrared images are invariant to illumination, but at the same time they are affected by a fixed-pattern noise inherent to this technology. The fixed-pattern is normally compensated by means of a non-uniformity correction method. Our study shows that the LBP approach is robust to the fixed-pattern noise, as well as to the presence of glasses. Not only no noise suppressing preprocessing is needed, but in fact if a non-uniformity correction method is applied, the image texture is amplified and the performance of the LBP degraded.

A recursive least square adaptive filter for nonuniformity correction of infrared image sequences

In this paper, an adaptive scene-based nonuniformity correction methodology for infrared image sequences is developed. The method estimates detector parameters and carry out the non-uniformity correction based on the recursive least square filter approach, with adaptive supervision. The key advantage of the method is based in its capacity for estimate detectors parameters, and then compensate for fixed-pattern noise in a frame by frame basics. The ability of the method to compensate for nonuniformity is demonstrated by employing several infrared video sequences obtained using two infrared cameras.

An Effective Reference-Free Performance Metric for Non-uniformity Correction Algorithms in Infrared Imaging System

In this paper we present the effective-roughness (ERo) index, a simple reference-free performance metric for nonuniformity correction methods that agrees with visual evaluations.

A cyclostationary analysis applied to detection and diagnosis of faults in helicopter gearboxes

In several cases the vibration signals generated by rotating machines can be modeled as cyclostationary processes. A cyclostationary process is defined as a non-stationary process which has a periodic time variation in some of its statistics, and which can be characterized in terms of its order of periodicity. This study is focused on the use of cyclic spectral analysis, as a tool to analyze second-order periodicity signals (SOP), such as, those who are generated by either localized or distributed defects in bearings. Cyclic spectral analysis mainly consists of the estimation of the random aspects as well as the periodic behavior of a vibration signal, based on estimation of the spectral correlation density. The usefulness of cyclic spectral analysis for the condition monitoring of bearings, is demonstrated in this paper, through the analysis of several sections of vibration data collected during an endurance test of one of the two main gearbox transmissions of a helicopter.

Simultaneous digital super-resolution and nonuniformity correction for infrared imaging systems

In this article, we present a novel algorithm to achieve simultaneous digital super-resolution and nonuniformity correction from a sequence of infrared images. We propose to use spatial regularization terms that exploit nonlocal means and the absence of spatial correlation between the scene and the nonuniformity noise sources. We derive an iterative optimization algorithm based on a gradient descent minimization strategy. Results from infrared image sequences corrupted with simulated and real fixed-pattern noise show a competitive performance compared with state-of-the-art methods. A qualitative analysis on the experimental results obtained with images from a variety of infrared cameras indicates that the proposed method provides super-resolution images with significantly less fixed-pattern noise.

Feature Analysis for the Classification of Volcanic Seismic Events Using Support Vector Machines

This paper shows a preliminary study to perform a pattern recognition process for seismic events of the Llaima volcano, one of the most active volcanoes in South America. 1622 classified events registered from the Llaima volcano were considered in this study, taken from 2009 to 2011. The events were divided in four classes: TREMOR (TR), LONG-PERIOD (LP), VOLCANO-TECTONICS (VT) and OTHERS (OT). All of them correspond to specific activities. TR and LP events, are related to magmatic fluid through the ducts: continuous flux correspond to TR and discrete flux to LP. VT events occurs when excess of the magmatic pressure provides enough energy for rock failure. The group of OT contains events not related to the three first volcanic classes. Many features extracted from de amplitude, the frequency and the phase of the events were used to characterize the different classes. A classifier step based on Support Vector Machines was implemented to evaluate the contribution of each feature to the classification. The paper shows the results of this process and gives insights for future works.

Quaternion correlation filters for illumination invariant face recognition

Illumination variations is one of the factors that causes the degradation of face recognition systems performance. The representation of face image features using the structure of quaternion numbers is a novel way to alleviate the illumination effects on face images. In this paper a comparison of different quaternion representations, based on verification and identification experiments, is presented. Four different face features approaches are used to construct quaternion representations. A quaternion correlation filter is used as similarity measure, allowing to process together all the information encapsulated in quaternion components. The experiment results confirms that using quaternion algebra together with existing face recognition techniques permits to obtain more discriminative and illumination invariant methods.

Block-recursive filtering for offset nonuniformity estimation in infrared imaging systems: Theory and implementation

In an earlier work, a recursive filter to compensate for the offset nonuniformity (NU) noise corrupting the output of infrared (IR) imaging system was presented. Such a filter was derived assuming an estimation time-window short enough so that the offset NU can be regarded as a constant corrupted by additive noise. In this paper, the assumption on the stationarity of the offset NU noise has been relaxed by characterizing the dynamics of the offset NU using a Gauss-Markov random process. Based upon this model, a recursive filter that uses blocks of IR data to estimate the offset NU has been derived. A rigorous theoretical analysis of the filter has been conducted in order to provide expressions for appropriately selecting the parameters of the filter. The ability of the block-recursive filter to compensate for NU noise has been tested using raw IR videos from different IR cameras. In addition, the filter has been implemented and tested on-line in a prototype spectrally tunable IR camera.

A RLS filter for nonuniformity and ghosting correction of infrared image sequences

In this paper, a technique to improve the convergence and to reduce the ghosting artifacts of a previously developed adaptive scene-based nonuniformity correction method is presented. The nonuniformity correction method estimates detector parameters based on the recursive least square filter approach. We propose, three parameters to reduce ghosting artifacts and to speed up the convergence of such method by using only the read-out data. The parameters proposed are based in identify global motion between consecutive frames as well as evaluate the main assumption used in the previous method in the uncertainty on the input infrared irradiance. The ability of the method to compensate for nonuniformity and reducing ghosting artifacts is demonstrated by employing several infrared video sequences obtained using two infrared cameras.

Filter design of composite trade-off filter with support regions to obtain invariant pattern recognition with defocused images

The discrimination capacity (DC) measures the ability of the filter in a pattern recognition problem to discriminate the target against other objects in the input scene. If the input scene is degraded by a defect of focus, then the DC is degraded and the pattern recognition process is worse. In this paper, we present a methodology based in the selection of ring frequency bands and in the design of the trade-off filters taking into account these frequencies to obtain several information channels. The information of all the channels is fused by means of the addition of all the channels and the geometric mean of them. Also individual channel analysis is shown. The influence on the DC and SNR of the added white noise in the input image is presented.