Polychromatic pattern recognition using the non-zero-order joint transform correlator with cross-correlation peak optimization

Abstract

A multi-channel polychromatic non-zero-order joint transform correlator (NOJTC) with minimum average cross-correlation energy is introduced. The system is implemented using a general JTC scheme with additional power-spectrum subtraction strategy and the design of an ideal reference function with a Lagrange multipliers technique. As the zero-order part is removed, the desired correlation peaks can be quite distinctive and sharp. The correlator also shows the advantage of better detection ability over monochromatic systems.     

Heteroassociative multiple-target tracking by fringe-adjusted joint transform correlation

A heteroassociative joint transform correlation (JTC) technique is proposed for recognizing and tracking multiple heteroassociative or dissimilar targets from gray-level image sequences by use of the concept of fringe-adjusted JTC and a multiple-target-detection algorithm. A fringe-adjusted JTC technique is used to ensure quantification of the similarities among several input images while it satisfies the equal-correlation-peak criterion. Tracking is accomplished by retrieval of the target motion information estimated from multiple consecutive image frames. An enhanced version of the fringe-adjusted filter is incorporated into the heteroassociative multiple-target-detection process to optimize the correlation performance. The feasibility of the proposed technique is tested by computer simulation with real infrared image data.     

Feature-extracted joint transform correlation

A new technique for real-time optical character recognition that uses a joint transform correlator is proposed. This technique employs feature-extracted patterns for the reference image to detect a wide range of characters in one step. The proposed technique significantly enhances the processing speed when compared with the presently available joint transform correlator architectures and shows feasibility for multichannel joint transform correlation.     

Phase-encoded joint transform correlator to reduce the influence of extraneous signals

A technique to reduce the influence of the dc and the complex-conjugate correlation signals that are inherent in a joint transform correlator is proposed. The key part of this technique is the use of a phase mask. The mask contributes only extraneous signals, so the influence of the signals is reduced. The desirable correlation signal is not affected, and its shape is retained. Computer simulations confirm the performance of the proposed phase-encoded joint transform correlator.     

Experimental verifications of a joint transform correlator using compressed reference images

Single-target and multiple-target detections by using a joint transform correlator (JTC) with compressed reference images are experimentally verified. Two high-contrast images with different spatial-frequency content are used as test scenes. Although an effect of the additive noise on detection performance of the proposed correlator is more severe than that of the compression, the experimental results confirm the feasibility of implementing the JTC with compressed reference images.     

Joint transform correlator for optical temporal signals

Correlation is considered to be a fundamental operation in the field of signal processing. The fact that this operation can be implemented optically in a relatively simple manner is an important advantage of utilizing optical systems for signal processing. The VanderLugt 4-f system and the joint transform correlator (JTC) are the two most popular configurations for performing a spatial correlation operation optically. So far the JTC architecture has been used for performing correlation between two images, which are illuminated by a quasi-monochromatic light source. We propose a generalization of the JTC that performs a correlation between two temporal optical signals.     

Wavelet filter for improving detection performance of compression-based joint transform correlator

A new method for improving fingerprint detections by a compression-based joint transform correlator (JTC) via wavelet filter is proposed. The simulation results show that the proposed method has advantages over the conventional compression-based JTC in that detection performance can be maximized to be higher than a classical JTC by using a smaller file size for the compressed-reference images.     

Implementation of a non-zero-order joint-transform correlator by use of phase-shifting techniques

The implementation of non-zero-order joint-transform correlators (JTC's) is presented. The zero-order spectra (i.e., the autocorrelation power spectra) are removed from the joint-transform power spectrum by use of phase-shifting techniques by which the output diffraction and input spatial domain can more efficiently be utilized. Applications of the phase-shifting techniques to both conventional JTC's and phase-transformed input JTC's (PJTC's) are discussed. Compared with the conventional JTC, the PJTC has the advantages of higher light efficiency, a better signal-to-clutter ratio, and the simplicity to realize phase shifting. We anticipate that the proposed non-zero-order JTC's should have a significant impact on the future development of more efficient JTC's.     

Performance improvement of a phase-shifting joint transform correlator by use of phase-iterative techniques

We propose a new technique to improve the performance of the joint transform correlator (JTC). In this technique we have applied the phase-iterative algorithm to a phase-shifting JTC (PSJTC). By doing so, we restrain the noise that is contained in the recovered phase of the joint transform power spectra for the input images with background and additive noise. In the case in which the input image is embedded in the input noise, we find that, by using the phase-iterative techniques with the PSJTC, one can get a higher cross-correlation peak and signal-to-noise ratio than with a PSJTC alone. From the computer-simulation results, one can conclude that the proposed algorithm successfully enhances PSJTC performance, especially for an input image with large noise.     

Generalized joint fractional fourier transform correlators: a compact approach

Fractional correlation was introduced recently. We generalize the architecture of a joint (Fourier) transform correlator (JTC) to achieve the joint fractional (Fourier) transform correlator (JFrTC) such that fractional correlation can be obtained. Here the Fourier transform in the JTC is replaced by the fractional Fourier transform, and four different JFrTC architectures can be implemented. The mathematical derivations for these JFrTC architectures are given, together with the simulation verifications. The JFrTC can provide a correlation signal similar to a delta function but with a small discrimination ratio, such that it is insensitive to additive noise. In a conventional JTC the distance between the two desired correlation signals at the output plane is fixed and depends on the distance between the input and the reference signals. However, with a given fractional order and an additional phase mask the separation distance between the two correlation signals at the output plane of a JFrTC can be larger or smaller than that of a JTC. This property is useful for the applications of real-time target tracking. Unlike in a previous approach [Appl. Opt. 36, 7402 (1997)], we need only two fractional Fourier transformations instead of three to achieve fractional correlation.     

Improved fingerprint identification with supervised filtering enhancement

An important step in the fingerprint identification system is the reliable extraction of distinct features from fingerprint images. Identification performance is directly related to the enhancement of fingerprint images during or after the enrollment phase. Among the various enhancement algorithms, artificial-intelligence-based feature-extraction techniques are attractive owing to their adaptive learning properties. We present a new supervised filtering technique that is based on a dynamic neural-network approach to develop a robust fingerprint enhancement algorithm. For pattern matching, a joint transform correlation (JTC) algorithm has been incorporated that offers high processing speed for real-time applications. Because the fringe-adjusted JTC algorithm has been found to yield a significantly better correlation output compared with alternate JTCs, we used this algorithm for the identification process. Test results are presented to verify the effectiveness of the proposed algorithm.     

Fourier-plane windowing in the binary joint transform correlator for multiple target detection

With recent advances in state-of-the-art spatial light modulators, the optical joint transform correlator (JTC) and the binary joint transform correlator (BJTC) are becoming practical signal-processing tools. The performance of these devices is limited by the difficulty of separating the cross correlation between the reference and the targets in the scene from signals resulting from cross correlations between objects in the target scene. One technique that reduces this problem is to use a sliding window in the Fourier plane as a convolution mask filter to set an adaptive binarization threshold. This suppresses the autocorrelation response and reduces the dynamic range of the Fourier-plane signal. This results in correlation performance improvement by a factor of 2 to 4. A mathematical model is developed to describe the windowing process for both the JTC and BJTC for the case in which the scene contains multiple targets and background clutter. The derivation of the windowing process is general and includes any spatial high-pass or bandpass filtering in the Fourier plane. The results are supported with experimental data.     

High-efficiency hybrid joint transform correlator with pre-processed input image and replicated binary spectrum array

Abstract

The performance of a hybrid joint transform correlator (JTC) is improved by the input image pre-processing method and our proposed spectrum binarizing and replicating technique. The input image pre-processing can enhance the correlation discrimination ability of a JTC system, and the replication of the binary spectrum under an appropriate condition can further substantially increase the light efficiency, the maximum correlation peak intensity, the peak sharpness and the signal-to-noise ratio of the system at the same time by the mechanics of constructive interference. Theoretical analyses are presented. Both the computer simulation and optical experiment verifying the effectiveness of this technique are also given.     

Multiple-target detection with use of a modified amplitude-modulated joint transform correlator

Multiple-target detection with a modified amplitude-modulated joint transform correlator is proposed. With this technique the joint power spectrum is first modified by the subtraction of the power spectrum of the input scene only and of the reference image from it; the resultant modified joint power spectrum is next multiplied by the amplitude-modulated filter function. The effect of noise in the input scene on the performance of the joint transform correlator is analyzed and quantified. This technique is found to deliver a correlation output and the capacity to accommodate noise in the input scene better than both the fringe-adjusted filter-based joint transform correlator and the modified fringe-adjusted filter-based joint transform correlator.     

Binary joint transform correlator based on differential processing of the joint transform power spectrum

A novel binary joint transform correlator with differential processing in the joint transform power spectrum is proposed. The new model of the binary joint transform correlator can eliminate the dc component and sharpen the correlation intensity peaks, hence increasing its discrimination ability but with a lower complexity of computation. Preliminary results are shown by computer simulations.     

Three-dimensional object recognition by fourier transform profilometry

An automatic method for three-dimensional (3-D) shape recognition is proposed. It combines the Fourier transform profilometry technique with a real-time recognition setup such as the joint transform correlator (JTC). A grating is projected onto the object surface resulting in a distorted grating pattern. Since this pattern carries information about the depth and the shape of the object, their comparison provides a method for recognizing 3-D objects in real time. A two-cycle JTC is used for this purpose. Experimental results demonstrate the theory and show the utility of the new proposed method.     

基于液晶光阀微分处理的实时联合变换相关器

提出了一种用液晶光阀实现对输入图像的微分预处理和对功率谱微分处理的实时联合变换相关器。两个液晶光阀均处于非线性变换态,分别在输入面和频谱面上实现对输入图像和联合功率谱的微分处理。微分功率谱经傅里叶透镜变换后得到相关输出。实验表明,该相关器采用全光方式不仅保证了实时性,而且锐化了相关峰;经该相关器微分后的图像,其相关性能明显优于常规联合变换相关器输出的结果。     

Novel real-time joint-transform correlation by use of acousto-optic heterodyning

To replace the film recording aspect of performing optical correlation, conventional real-time joint-transform correlation (JTC) optical systems make use of a spatial light modulator (SLM) located in the Fourier plane to record the joint-transform power spectrum (JPS) to achieve real-time processing. The use of an SLM in the Fourier plane, however, is a major drawback in these systems because SLMs are limited in resolution, phase uniformity, and contrast ratio, which are, therefore, not desirable for robust applications. We propose a hybrid (optical/electronic) processing technique to achieve real-time joint-transform correlation. The technique employs acousto-optic heterodyning scanning. The proposed real-time JTC system does not require an SLM at the Fourier plane as in other real-time JTC systems. This departure from the conventional scheme is extremely important as the proposed approach does not depend on SLM issues. We develop the theory of the technique and substantiate it with experimental results.     

Detection performance of wavelet-based joint transform correlation

The detection performance of a wavelet-based joint transform correlator (JTC) is studied by use of two types of images with different spatial-frequency contents and contrast. The simulation results show that, in comparison with an amplitude-modulated JTC, the performance for intraclass pattern recognition can be optimized by using a single wavelet filter.     

All-optical subtracted joint transform correlator with a holographic interferometer

A new, to our knowledge, subtracted joint transform correlator (SJTC) is proposed that has no digital processing in a computer. All processing for obtaining correlation signals between an object and multiple reference patterns were treated optically by use of a joint transform correlator with a holographic interferometer similar to the Mach-Zehnder one. The joint power spectrum of the reference patterns was subtracted from that of the input image (the object pattern plus the reference patterns), and the spurious correlation signals between the different reference patterns were removed. Because of the optical parallel computations of the Fourier spectra and the subtraction, the real-time SJTC is possible to achieve by use of only an optical system. An experimental arrangement of the system and system performances of the shift-invariant characteristics and discriminability are described. The results show the good performance of this system.