Kostas N. Plataniotis

Self-adaptive algorithm of impulsive noise reduction in color images

In this paper a new approach to the problem of impulsive noise reduction in color images is presented. The basic idea behind the new image filtering technique is the maximization of the similarities between pixels in a predefined filtering window. The improvement introduced to this technique lies in the adaptive establishing of parameters of the similarity function and causes that the new filter adapts itself to the fraction of corrupted image pixels. The new method preserves edges, corners and fine image details, is relatively fast and easy to implement. The results show that the proposed method outperforms most of the basic algorithms for the reduction of impulsive noise in color images.

On solving the face recognition problem with one training sample per subject

The lack of adequate training samples and the considerable variations observed in the available image collections due to aging, illumination and pose variations are the two key technical barriers that appearance-based face recognition solutions have to overcome. It is a well-documented fact that their performance deteriorates rapidly when the number of training samples is smaller than the dimensionality of the image space. This is especially true for face recognition applications where only one training sample per subject is available. In this paper, a recognition framework based on the concept of the so-called generic learning is introduced as an attempt to boost the performance of traditional appearance-based recognition solutions in the one training sample application scenario. Different from contemporary approaches, the proposed solution learns the intrinsic properties of the subjects to be recognized using a generic training database which consists of images from subjects other than those under consideration. Many state-of-the-art face recognition solutions can be readily integrated in the proposed framework. A novel multi-learner framework is also proposed to further boost recognition performance. Extensive experimentation reported in the paper suggests that the proposed framework provides a comprehensive solution and achieves lower error recognition rate when considered in the context of one training sample face recognition problem.

A multichannel order-statistic technique for cDNA microarray image processing

This paper introduces an automated image processing procedure capable of processing complementary deoxyribonucleic acid (cDNA) microarray images. Microarray data is contaminated by noise and suffers from broken edges and visual artifacts. Without the utilization of a filter, subsequent tasks such as spot identification and gene expression determination cannot be completed. By employing, in a unique cascade processing cycle, nonlinear filtering solutions based on robust order statistics, the procedure: 1) removes both background and high-frequency corrupting noise and 2) correctly identifies edges and spots in cDNA microarray data. The proposed solution operates directly on the microarray data, does not rely on explicit data normalization or spot separation preprocessing, and operates in a robust manner without using heuristically determined design parameters. Other routine microarray processing operations such as shape manipulations and grid adjustments can be used in conjunction with the developed solution in the processing pipeline. Experimentation reported in this paper indicates that the proposed solution yields excellent performance by removing noise and enhancing spot location determination.

Automatic location and tracking of the facial region in color video sequences

A novel technique is introduced to locate and track the facial area in videophone-type sequences. The proposed method essentially consists of two components: (i) a color processing unit, and (ii) a knowledge-based shape and color analysis module. The color processing component utilizes the distribution of skin-tones in the HSV color space to obtain an initial set of candidate regions or objects. The second component in the segmentation scheme, that is, the shape and color analysis module is used to correctly identify and select the facial region in the case where more than one object has been extracted. A number of fuzzy membership functions are devised to provide information about each object’s shape, orientation, location and average hue. An aggregation operator finally combines these measures and correctly selects the facial area. The suggested approach is robust with regard to di⁄erent skin types, and various types of object or background motion within the scene. Furthermore, the algorithm can be implemented at a low computational complexity due to the binary nature of the operations involved. Experimental results are presented for a series of CIF and QCIF video sequences. ( 1999 Elsevier Science B.V. All rights reserved.

Towards automatic redeye effect removal

The redeye effect is typically formed in amateur photographs taken with a built-in camera flash. Analysis of the available techniques and products indicates that their efficiency in correcting this artifact is limited and their performance is inconsistent. In this work we propose a user friendly solution, which could be used to restore amateur photographs. In the proposed method the redeye effect is detected using a skin detection module and eye colors are restored using morphological image processing. The new method is computationally efficient, robust to parameter settings and versatile, as it can work in conjunction with a number of skin detection methods.

Face recognition with biometric encryption for privacy-enhancing self-exclusion

Face recognition has been employed in various security-related applications such as surveillance, mugshot identification, e-passport, and access control. Despite its recent advancements, privacy concern is one of several issues preventing its wider deployment. In this paper, we address the privacy concern for a self-exclusion scenario of face recognition, through combining face recognition with a simple biometric encryption scheme called helper data system. The combined system is described in detail with focus on the key binding procedure. Experiments are carried out on the CMU PIE face database. The experimental results demonstrate that in the proposed system, the biometric encryption module tends to significantly reduce the false acceptance rate while increasing the false rejection rate.

Adaptive multichannel filters for colour image processing

Abstract This paper addresses the problem of noise suppression for multichannel data, such as colour images. The proposed filters utilize adaptive data dependent nonparametric techniques. Simulation results indicate that the new filters suppress impulsive as well as Gaussian noise and preserve edges and details.

Parallelization and performance of 3D ultrasound imaging beamforming algorithms on modern clusters

Recently there has been a lot of interest in improving the infrastructure used in medical applications. In particular, there is renewed interest on non-invasive, high-resolution diagnostic methods. One such method is digital, 3D ultrasound medical imaging. Current state-of-the-art ultrasound systems use specialized hardware for performing advanced processing of input data to improve the quality of the generated images. Such systems are limited in their capabilities by the underlying computing architecture and they tend to be expensive due to the specialized nature of the solutions they employ.Our goal in this work is twofold: (i) To understand the behavior of this class of emerging medical applications in order to provide an efficient parallel implementation and (ii) to introduce a new benchmark for parallel computer architectures from a novel and important class of applications. We address the limitations faced by modern ultrasound systems by investigating how all processing required by advanced beamforming algorithms can be performed on modern clusters of high-end PCs connected with low-latency, high-bandwidth system area networks. We investigate the computational characteristics of a state-of-the-art algorithm and demonstrate that todays commodity architectures are capable of providing almost-real-time performance without compromising image quality significantly.

On the distance function approach to color image enhancement

A new class of image processing filters is introduced and analyzed in this paper. The new filters utilize fuzzy measures applied to image pixels connected by digital paths. The performance of the proposed filters is compared to the performance of commonly used filters, such as the vector median, under a variety of performance criteria. It is shown that the proposed filters are better able to suppress impulsive and Gaussian noise than the existing techniques. Also, they are robust to inaccuracies in parameter settings.

Fast Modified Vector Median Filter

A new filtering approach designed to eliminate impulsive noise in color images, while preserving fine image details is presented in this paper. The computational complexity of the new filter is significantly lower than that of the Vector Median Filter. The comparison shows that the new filter outperforms the VMF, as well as other standard procedures used in color image processing, when the impulse noise is to be eliminated.