Distortion-invariant filter for nonoverlapping noise

Abstract

A new heuristic filter based on the optimum filter for disjoint noise developed by Javidi and Wang [J. Opt. Soc. Am. A 11, 2604 (1995)] is presented. In this new filter a number of optimum filters built from single training images are combined linearly by use of the synthetic discriminant function (SDF) approach into a distortion-invariant filter for disjoint noise. Like the traditional SDF approach, this summation technique makes it possible to control the height of the correlation peak easily, for example, if a uniform filter response is needed. The filter is compared with the distortion-invariant version of the optimum filter on images with low contrast and high levels of nonoverlapping clutter. The new filter shows good results, demonstrating that it is, with very simple heuristic methods, possible to improve the performance of distortion-invariant filters for nonoverlapping noise.

Distortion-Invariant Pattern Recognition with Adaptive Correlation Filters

“…Various composite optimum correlation filters for recognition of geometrically distorted objects embedded in a nonoverlapping background have been proposed (Chan et al, 2000;Sjöberg & Noharet, 1998). However, there are no correlation-based methods for detection and localization of geometrically distorted objects in blurred and noisy scenes.…”

Nonlinear synthetic discriminant function filters for illumination-invariant pattern recognition

“…16 Recently, composite optimum correlation filters for recognition of geometrically distorted objects embedded into a nonoverlapping background were also proposed. [17][18][19] Another fruitful approach to distortion-invariant pattern recognition is based on adaptive filters. [20][21][22] According to this concept, we are looking for a filter with good performance characteristics for a given set of observed scenes, i.e., with a fixed set of patterns and backgrounds to be rejected, rather than for a filter with average performance parameters over an ensemble of images.…”

Quantised phase filters with binary low and high pass amplitude response: the effect of quantisation for scale and rotation input distortions

“…These include the use of phase-only filters (POFs) [3,4], circular harmonic expansion (CHE) functions [5], rotationally multiplexed holograms [6], synthetic discriminant functions (SDFs) [7,8], minimum average correlation energy (MACE) filters [9], phase with constrained amplitude filters (PCMF) [10], the Wiener filter [11,12], and the ternary phase-amplitude filter (TPAF) [13][14][15].…”