Nonlinear synthetic discriminant function filters for illumination-invariant pattern recognition

Abstract

Novel nonlinear adaptive composite filters for illuminationinvariant pattern recognition are presented. Pattern recognition is carried out with space-variant nonlinear correlation. The information about objects to be recognized, false objects, and a background to be rejected is utilized in an iterative training procedure to design a nonlinear adaptive correlation filter with a given discrimination capability. The designed filter during recognition process adapts its parameters to local statistics of the input image. Computer simulation results obtained with the proposed filters in nonuniformly illuminated test scenes are discussed and compared with those of linear composite correlation filters with respect to discrimination capability, robustness to input additive and impulsive noise, and tolerance to small geometric image distortions.

Morphological Correlation for Robust Image Recognition

“…One can show that the morphological correlation yields the value u for objects belonging to the training set [8].…”

“…On the other hand, nonlinear filters have been used for detection of noisy objects [7]; these kinds of filters are more robust in scenes corrupted by non-Gaussian noise. Moreover, it is possible to design nonlinear composite filters to achieve distortion invariance [8]. In this paper we propose the use of morphological correlation combined with nonlinear filters for robust image recognition.…”

Rotation-Invariant Nonlinear Filters Design

“…Recently, nonlinear composite filters for distortion-invariant pattern recognition were introduced [10]. The filters are designed as a logical combination of binary objects.…”

Distortion-invariant pattern recognition with nonlinear correlation filters

“…9,10 Recently, novel nonlinear composite filters for illumination-invariant pattern recognition were proposed. 11 The filters are designed by applying logical operations among a set of training objects. Various objects to be recognized and rejected can be incorporated in the template synthesis.…”

“…[1][2][3][4][5][6][7]10,11 Several reasons make difficult the recognition task; one of such reasons is geometrical distortions of objects to be recognized. To overcome this problem, composite filters were introduced; these filters incorporate information from various objects to synthesize a template.…”