Wavelet-transform-based composite filters for invariant pattern recognition

A wavelet-transformation-based optical processor for performing invariant pattern recognition is suggested. It contains a composite filter that consists of several wavelet daughter functions derived from the reference object. The intensity of the correlation peak is determined to be invariant to various deformations of the reference object. Computer simulations show explicitly the promising capability of the new technique. Laboratory experimental results are given.     

Continuous-scale-invariant pattern recognition with adaptive-wavelet-matched filters

We designed and implemented a composite-wavelet-matched filter that is invariant to continuous-scale changes of the input and is useful for correlation-based pattern recognition of objects whose size is not known exactly. We optimize the adaptive wavelet to extract sparse image features and use the scale-space analysis to determine the wavelet scale for the scale invariance. Experimental results obtained by use of a programmable optical correlator with three liquid-crystal spatial light modulators are demonstrated.     

Distortion-invariant pattern recognition with Fourier-plane nonlinear filters

The use of nonlinear techniques in the Fourier plane of pattern-recognition correlators can improve the correlators' performance in terms of discrimination against objects similar to the target object, correlation-peak sharpness, and correlation noise robustness. Additionally, filter designs have been proposed that provide the linear correlator with invariance properties with respect to input-signal distortions and rotations. We propose simple modifications to presently known distortion-invariant correlator filters that enable these filter designs to be used in a nonlinear correlator architecture. These Fourier-plane nonlinear filters can be implemented electronically, or they may be implemented optically with a nonlinear joint transform correlator. Extensive simulation results are presented that illustrate the performance enhancements that are gained by the unification of nonlinear techniques with these filter designs.     

Color encoding for polychromatic single-channel optical pattern recognition

The common multichannel system for recognizing colored images is replaced by a color-encoded single-channel system. Amethod inspired by the Munsell color system is used for encoding the different colors as phase and amplitude functions. It is shown that for many practical cases the phase information part of the color code is sufficient for obtaining good results. An implementation based on a liquid-crystal television panel that works in a phase-modulation mode is suggested. Computer simulations that demonstrate the capabilities of the suggested method are given as well as a comparison with previously published multichannel performance.     

Projection-slice synthetic discriminant functions for optical pattern recognition

The projection-slice synthetic discriminant function (PSDF) filter is introduced and proposed for distortion-invariant pattern-recognition applications. The projection-slice theorem, often used in tomographic applications for medical imaging, is utilized to implement a distortion-invariant filter. Taking M projections from one training image and combining them with (N - 1)M projections taken from another N - 1 training image accomplishes this. With the projection-slice theorem, each set of these M projections can be represented as M one-dimensional slices of the two-dimensional Fourier transform of the particular training image. Therefore, the PSDF filter has the advantage of matching each of the training images with at least M slices of their respective Fourier transforms. This filter is theoretically analyzed, numerically simulated, and experimentally implemented and tested to verify the simulation results. These tests show that the PSDF filter significantly outperforms the matched-filter and the basic synthetic discriminant function technique for the particular images used.     

Hybrid incoherent optical pattern recognition system

A pattern recognition system that uses incoherent spatial filtering to recognize images directly from a narrowband phosphor television monitor is described. Images of real objects are captured with a television camera. These images are then edge-enhanced electronically and displayed on the TV monitor. The monitor output is used directly as the input to a holographic correlator. An optical multichannel analyzer at the correlation plane is used to analyze the shape of the correlation function and to determine the position of its peak. Experimental results agree well with theory. Concepts for handling rotation, aspect angle, and scale variations of the input are discussed.     

Cascaded linear shift-invariant processors in optical pattern recognition

We study a cascade of linear shift-invariant processing modules (correlators), each augmented with a nonlinear threshold as a means to increase the performance of high-speed optical pattern recognition. This configuration is a special class of multilayer, feed-forward neural networks and has been proposed in the literature as a relatively fast best-guess classifier. However, it seems that, although cascaded correlation has been proposed in a number of specific pattern recognition problems, the importance of the configuration has been largely overlooked. We prove that the cascaded architecture is the exact structure that must be adopted if a multilayer feed-forward neural network is trained to produce a shift-invariant output. In contrast with more generalized multilayer networks, the approach is easily implemented in practice with optical techniques and is therefore ideally suited to the high-speed analysis of large images. We have trained a digital model of the system using a modified backpropagation algorithm with optimization using simulated annealing techniques. The resulting cascade has been applied to a defect recognition problem in the canning industry as a benchmark for comparison against a standard linear correlation filter, the minimum average correlation energy (MACE) filter. We show that the nonlinear performance of the cascade is a significant improvement over that of the linear MACE filter in this case.     

Rotationally invariant pattern recognition by use of linear and nonlinear cascaded filters

We discuss the merits of using single-layer (linear and nonlinear) and multiple-layer (nonlinear) filters for rotationally invariant and noise-tolerant pattern recognition. The capability of each approach is considered with reference to a two-class, rotation-invariant, character recognition problem. The minimum average correlation energy (MACE) filter is a linear filter that is generally accepted to be optimal for detecting signals that are free from noise. Here it is found that an optimized MACE filter cannot differentiate between the characters E and F in a rotation-invariant manner. We have found, however, that this task is possible when a single optimized linear filter is used to achieve the required response when a nonlinear threshold function is included after the filter. We show that this structure can be cascaded to form a multiple-layer, cascaded filter and that the capability of such a system is enhanced by its increased noise tolerance in the character recognition problem. Finally, we show the capability of a two-layer cascade as a means to detect different species of bacteria in images obtained from a phase-contrast microscope.     

Selection of optimal filters for multispectral imaging

Preece and Claridge [IEEE Trans. Pattern Anal. Mach. Intell. 26, 913 (2004)] have proposed a technique for selecting filters for the maximally accurate recovery of object parameters such as chromophore concentrations from a multispectral image of an object. Their selection criteria are derived from an analysis of a model of light propagation in the object and take into account both errors in the modeling process and errors in the image acquisition process, as well as the inherent behavior and structure of the model. We investigate their method on simulated image data and show that filters selected according to their criteria are demonstrably superior to other choices.     

Size-speed trade-off in optical switching elements

The basic building blocks of an interconnection network are the switching elements. We examine two optical implementations for a basic switching element: (1) ferroelectric liquid-crystal light valves and (2) Fabry-Perot étalons. For these two examples we report a trade-off between the size, i.e., the number of input and output channels, and the switching speed. We speculate that it may be a general property of optical switching elements that size and speed cannot be optimized simultaneously.     

Generality of matched filtering and minimum Euclidean distance projection for optical pattern recognition

Matched filtering followed by a minimum Euclidean distance projection onto realizable filter values was previously shown to optimize the signal-to-noise ratio for single training images in optical correlation pattern recognition. The algorithm is now shown to solve the combination of (1) standard statistical pattern-recognition metrics with multiple training images, (2) additive input noise of known power spectral density and also additive detection noise that is irreducible by the filter, (3) the building of the filter on arbitrary subsets of the complex unit disk, and (4) the use of observable correlator outputs only. The criteria include the Fisher ratio, the Bayes error and Bayes cost, the Chernoff and Bhattacharyya bounds, the population entropy and expected information, versions of signal-to-noise ratio that use other than second power in their norm, and the area under the receiver operating characteristic curve. Different criteria are optimized by different complex scalar weights.     

Application of correlation filters for texture recognition

We propose a new statistical method to design spatial filters to recognize and to discriminate between various textures. Unlike existing correlation filters, the proposed filters are not meant to recognize specific shapes or objects. Rather, they discriminate between textures such as terrains, background surfaces, and random image fields. The filters do not require any on-line statistical computations for extracting texture information. Therefore optical (or digital) correlators can be used for fast real-time texture recognition without segmentation. The procedure is based on the assumption that textures can be modeled as stationary random processes over limited regions of an image. The optimum filter coefficients are determined by use of eigenvector analysis. Several examples are given to illustrate the proposed scheme.     

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.     

Nonlinear rotation-invariant pattern recognition by use of the optical morphological correlation

We introduce a modification of the nonlinear morphological correlation for optical rotation-invariant pattern recognition. The high selectivity of the morphological correlation is conserved compared with standard linear correlation. The operation performs the common morphological correlation by extraction of the information by means of a circular-harmonic component of a reference. In spite of some loss of information good discrimination is obtained, especially for detecting images with a high degree of resemblance. Computer simulations are presented, as well as optical experiments implemented with a joint transform correlator.     

Narrowband optical filters suitable for various applications in optical communications

We propose three types of narrowband optical filters based on a Fox-Smith resonator. We demonstrate that by choosing the appropriate combination of coating materials on each prism facet, one can design either a high reflectance or a high transmittance optical filter, suitable for low bit rate optical communication applications with International Telecommunication Union-Telecommunication Standardization Sector (ITU-T) standards. We also show the possibility of designing an optical filter having a desirable finite reflectance/transmittance ratio with simultaneous peaks at ITU-T standard wavelengths. Such filters could be suitable for wavelength tuning applications.     

Reliable fabrication technologies for optical resonant filters

A reliable technology for optical resonant filters can be useful for optoelectronics technologies, optical communications systems, and biomedical applications. Positioning the resonance at a specific target wavelength is challenging because of the sensitivity of placing a narrow spectral filtering window tothe error factors that are associated with the fabrication and characterization processes. We describe and prove experimentally two fabrication approaches to overcome this challenge.     

Noise robustness of nonlinear filters for image recognition

We analyze the performance of the Fourier plane nonlinear filters in terms of signal-to-noise ratio (SNR). We obtain a range of nonlinearities for which SNR is robust to the variations in input-noise bandwidth. This is shown both by analytical estimates of the SNR for nonlinear filters and by experimental simulations. Specifically, we analyze the SNR when Fourier plane nonlinearity is applied to the input signal. Using the Karhunen-Loève series expansion of the noise process, we obtain precise analytic expressions of the SNR for Fourier plane nonlinear filters in the presence of various types of additive-noise processes. We find a range of nonlinearities that need to be applied that keep the output SNR of the filter stable relative to changes in the noise bandwidth.     

Nonlinear cascaded correlation processes to improve the performances of automatic spatial-frequency-selective filters in pattern recognition

A recognition process consisting of two cascaded correlation stages with a sigmoid nonlinearity applied in the first correlation plane is investigated. The filters are computed to give prespecified central correlation amplitudes in the second correlation plane when inputs are reference images. It is also desired that the second correlation amplitudes with the training images should minimize the cost function of the automatic spatial-frequency selection algorithm to reduce distortion sensitivity and to improve the performance of the filters. Filter computation methods are given, and it is shown why two such correlation processes may improve the correlation performance. Numerical simulations are described and compared with the one-stage correlation system that works with the automatic spatial-frequency selection filter.     

Nonlinear techniques in optical pattern recognition: introduction by the feature editors

This feature of Applied Optics is dedicated to research in optical pattern-recognition techniques that utilize nonlinear devices and algorithms. Nonlinearities are employed in a variety of ways in optical pattern recognition and play an important role in both hardware implementation and software development for optical pattern-recognition systems.