Distance Classifier Correlation Filter Research Articles

Multiclass Pattern Recognition of Facial Images using Correlation Filters

Pattern Recognition comes naturally to humans and there are many pattern recognition tasks which humans can perform admirably well. However, human pattern recognition cannot compete with machine speed when the number of classes to be recognized becomes tremendously large. In this paper, we analyze the effectiveness of correlation filters for pattern classification problems. We have used Distance Classifier Correlation Filter (DCCF) for pattern classification of facial images. Two essential qualities of a correlation filter are distortion tolerance and discrimination ability. DCCF transposes the feature space in such a way that the images belonging to the same class gets closer and the images from different class moves far apart; thereby increasing the distortion tolerance and the discrimination ability. The results obtained demonstrate the effectiveness of the approach for face recognition applications.

Trends in correlation-based pattern recognition and tracking in forward-looking infrared imagery.

In this paper, we review the recent trends and advancements on correlation-based pattern recognition and tracking in forward-looking infrared (FLIR) imagery. In particular, we discuss matched filter-based correlation techniques for target detection and tracking which are widely used for various real time applications. We analyze and present test results involving recently reported matched filters such as the maximum average correlation height (MACH) filter and its variants, and distance classifier correlation filter (DCCF) and its variants. Test results are presented for both single/multiple target detection and tracking using various real-life FLIR image sequences.

New two-stage correlation-based approach for target detection and tracking in forward-looking infrared imagery using filters based on extended maximum average correlation height and polynomial distance classifier correlation

A novel approach is proposed to recognize and track multiple identical and/or dissimilar targets in forward-looking infrared (FLIR) image sequences using a combination of an extended maximum average correlation height (EMACH) filter and polynomial distance classifier correlation filter (PDCCF). The EMACH filter and PDCCF are trained a priori using representative training images of targets with expected size and orientation variations. In the first step, the input scene is correlated with all EMACH filters (one for each desired or expected target class). Based on the regions with higher correlation peak values in the combined correlation output, a sufficient number of regions of interest (ROIs) are selected from the input scene. In the second step, a PDCCF is applied to these ROIs to identify target types and reject clutter and background. Moving-target detection and tracking is accomplished by applying this technique independently to all incoming image frames. Independent tracking of target(s) from one frame to the other allows the system to handle complicated situations such as a target disappearing in a few frames and then reappearing in later frames. This method yields robust performance for challenging FLIR imagery in terms of accurate detection and classification as well as tracking of the targets.

Target detection, classification, and tracking using a maximum average correlation height and polynomial distance classification correlation filter combination

Simultaneous detection and classification of single/multiple identical and dissimilar targets is very important in automatic target recognition applications. A new approach is proposed for this purpose using a combination of maximum average correlation height (MACH) filter and polynomial distance classifier correlation filter (PDCCF). In this technique, full-resolution MACH filters are applied to the input scene, and the regions of interest (ROIs) containing the probable targets are selected from the input scene based on the ROIs with higher-correlation peak values in the correlation output. Then a multiclass PDCCF is applied to these ROIs to identify target types and reject clutters and/or backgrounds. To increase the robustness of the proposed technique, multiple filters are formulated for multiple ranges of target size and/or orientation variations. The simulation results using real-life imagery indicate the effectiveness of the proposed technique for target detection and classification in the presence of distortion, clutter, and other artifacts.

Face recognition subject to variations in facial expression, illumination and pose using correlation filters

In this paper, we have selected some recent advanced correlation filters: minimum average correlation filter (MACE), unconstrained MACE filter (UMACE), phase-only unconstrained MACE filter (POUMACE), distance-classifier correlation filter (DCCF) [B.V.K. Vijaya Kumar, D. Casasent, A. Mahalanobis, Distance-classifier correlation filters for multiclass target recognition. Appl. Opt. 35 (1996) 3127-3133] and minimax distance transform correlation filter (MDTC) and used them to test recognition performance in different situations involving variations in facial expression, illumination conditions and head pose. The paper introduces the first application of correlation filter classifiers to facial images subject to head pose variations. It also demonstrates that it is possible to obtain illumination invariance without using any training images for this purpose. A comparison of MDTC with traditional discriminant learning methods (e.g., KPCA [Scholikopf, B., Smola, A., Muller, K.R., Nonlinear component analysis as a kernel eigenvalue problem. Neural Comput., 10 (1999) 1299-1319], IPCA [16], GDA [Baudat, G., Anouar, F., Generalized discriminant analysis using a kernel approach. Neural Comput., 12 (2000) 2385-2404], R-KDA [Lu, J., Plataniotis, K., Venetsanopoulos, A. Regularization studies of linear discriminant analysis in small sample size scenarios with application to face recognition. Pattern Recogn. Lett., 26 (2) (2005) 181-191)] is also presented. The paper shows that correlation filter classifiers, a relatively unheralded model-based approach, have a greater robustness and accuracy than traditional appearance-based methods (such as PCA). Overall, the POUMACE filter provided the best choice for facial matching. It achieved 100% accuracy on the publicly available CMU facial expression database and the Yale frontal face illumination database, and slightly less in the head pose experiments.

Polynomial distance classifier correlation filter for pattern recognition.

We introduce what is to our knowledge a new nonlinear shift-invariant classifier called the polynomial distance classifier correlation filter (PDCCF). The underlying theory extends the original linear distance classifier correlation filter [Appl. Opt. 35, 3127 (1996)] to include nonlinear functions of the input pattern. This new filter provides a framework (for combining different classification filters) that takes advantage of the individual filter strengths. In this new filter design, all filters are optimized jointly. We demonstrate the advantage of the new PDCCF method using simulated and real multi-class synthetic aperture radar images.

Two class minimax distance transform correlation filter.

A new correlation filter formulation (that we refer to as the minimax distance transform correlation filter (MDTCF) is presented that minimizes the average squared distance from the filtered desired (or true-) class training images to a filtered reference image while maximizing the mean squared distance of the filtered undesired (or false-) class training images to this filtered reference image. This approach increases the separation between the false-class correlation outputs and the true-class correlation outputs. Classification can be performned using the squared distance of a filtered test image to the chosen filtered reference image. We show that the previously introduced distance classifier correlation filter (DCCF) is similar to a special case of MDTCF. We also examine the differences between the DCCF and the MDTCF, and show that MDTCF can offer increased discrimination performance. Also, MDTCF performance is evaluated on two different face databases.

Intraclass and between-class training-image registration for correlation-filter synthesis

Correlation filters have traditionally been designed without much attention given to the issue of the training images within a class or the relative spatial position between classes. We examine the impact of training-set registration on correlation-filter performance and develop techniques for centering the training images from a class that result in improved performance. We also show that it is beneficial to adjust the spatial position of the classes relative to one another. Although the proposed techniques are relevant for many types of correlation filter, we limit our discussion to algorithms for the maximum average correlation height filter and the distance classifier correlation filter.

Circular optimal trade-off and distance-classifier correlation filters

We use circular versions of advanced distortion-invariant filters such as optimal trade-off synthetic discriminant function and distance classifier correlation filters to obtain rotation invariance with an optical correlator. The filter noise performance is compared using a common measure of probability of error because the filters have different characteristics. The filters are real-valued so they can be implemented on a variety of SLMs. The circular symmetry of the filters significantly decreases their computational requirement.

Distance-classifier correlation filters for multiclass target recognition

We describe a correlation-based distance-classifier scheme for the recognition and the classification of multiple classes. The underlying theory uses shift-invariant filters to compute distances between the input image and ideal references under an optimum transformation. The original distance-classifier correlation filter was developed for a two-class problem. We introduce a distance-classifier correlation filter that simultaneously considers multiple classes, and we show that the earlier two-class formulation is a special case of the classifier presented. Initial results are presented to demonstrate the discrimination- and distortion-tolerance capabilities of the proposed filter.