In our former research on YUV decomposition for color image recognition with joint transform correlator, the training images are displayed at the center position. However, the total side lobe energy may not be the minimum. In order to solve the problem, three components are rotated from -50° to 50° in steps of 2° and shifted from -3 to 3 pixels in both of vertical and the horizontal directions to yield shifted training images. 3 reference functions with lower side lobe energy on the output plane are yielded by using the method of minimum average cross correlation energy.
A construction of the optoelectronic system with binary power spectrum is presented for target recognition. In the
beginning, the minimum average correlation energy method is used to yield the reference function. Then, the multi-level
quantized reference function is implemented at the input plane of the liquid crystal spatial light modulator in the joint
transform correlator . Numerical result is presented.
We propose a construction of the optoelectronic pattern recognition system with binary power spectrum for target
recognition, and apply liquid crystal spatial light modulators to this system. In addition, we utilize the minimum average
correlation energy method and multi-level quantized reference function. The constructed reference function could be
implemented at the input plane of the liquid crystal spatial light modulator. Numerical result is presented.
The CIELAB standard color vision model instead of the traditional RGB color model is utilized for polychromatic
pattern recognition. The image encoding technique is introduced. The joint transform correlator is set to be the optical
configuration. To achieve the distortion invariance in discrimination processes, we have used the minimum average
correlation energy approach to yield sharp correlation peak. From the numerical results, it is found that the recognition
ability based on CIELAB color specification system is accepted.
KEYWORDS: RGB color model, Visual process modeling, Computer programming, Pattern recognition, Distortion, Optical correlators, Human vision and color perception, Color vision, Target recognition, Image processing
In this paper, we use the ATD color human vision model instead of the traditional RGB color model for
polychromatic pattern recognition. Here we utilize the Mach-Zehnder joint transform correlator to be the optical pattern
discrimination configuration. The ATD color human vision model is proposed to be the interpretation of human's cone
mechanisms. The ATD represents the nonopponent achromatic system, the tritanopic system, and the deuteranopic
system, respectively. It is more close to human eye's vision. Besides, we also use the minimum average correlation
energy approach based on the Lagrange multipliers and image encoding technique to yield sharp correlation peak and to
achieve the distortion invariance. The image encoding method can reduce the requirement of pixels number in
RLCSLM on output plane effectively. Therefore, an encoded image recombined by the three vectors of ATD has been
utilized in our input plane. The minimum average correlation energy approach is designed to deal with various
distortions and to reduce the correlation sidelobe intensity. From these results, we discover that the recognition ability
based on ATD vector model is better than that based on RGB color model generally. Subsequently, we choose one
target from the 25 images set to estimate the discrimination ability in rotated distortion and noisy distortion. From the
numerical results we realize that the recognition ability based on ATD color vision model is accepted.
A novel two-channel single-output joint transform correlator system with the Mach-Zehnder configuration using
encoding technique based on HSV color space for color pattern recognition is introduced. The large zero order term can
be removed directly by the Stokes relations in only one step in this structure. By the image encoding technique, the size
of the liquid crystal spatial light modulators will be smaller with interlaced rearrangement of hue and saturation color
components. Furthermore, the utilization of Lagrange multipliers to synthesize the reference image for reducing the
correlation sidelobes is also studied. The computer numerical results are presented to verify the performance of the
proposed system.
A novel nonzero-order joint transform correlator system with the Mach-Zehnder configuration for polychromatic pattern
recognition is presented. To remove zero order term is a very important thing for optical pattern recognition. We cleverly
utilize the Stokes relations to remove the large zero order term directly in only one step in this structure. In the proposed
technique, we investigate color pattern recognition with YIQ channels involving in-plane distortion such as rotations.
Furthermore, the utilization of minimum average correlation energy to generate the composite reference image for
enhancing the correlation output is also studied. The computer simulation results are presented to show the validity of the
proposed technique.
In this article, we apply a novel method to polychromatic pattern recognition with a Mach-Zehnder nonzero order joint transform correlator. It is achieved by using the special arrangement of separated polychromatic components of objects and reference images and Lagrange multipliers technique to synthesize a reference function in the input spatial domain. The purpose is to sharpen the correlation peak and to reduce correlation sidelobes. Besides, we cleverly utilize the Stokes relations to remove the large zero order term directly in one step in this structure. Simulation results are presented to illustrate the improvements.
We present a novel color pattern recognition technology based on non-zero order joint transform correlator (NOJTC) system in this paper. In this method, each of the color target image is transformed to a grayscale image by using encoding technique. We also use minimum average correlation energy (MACE) approach to design an optimized synthetic reference function. When the input plane is gray-scaled and monochromatic, the function can be displayed in the liquid crystal spatial light modulator (LCSLM) to achieve real-time operation. Furthermore, we apply a joint transform power spectrum (JTPS) subtraction method to remove the zero-order terms and the desired peaks can be easily detected.
A novel phase-encoded amplitude-modulated joint transform correlator (PAJTC) technique is presented, which yields superior output correlation signals and better noise robustness than a classical JTC. In the proposed PAJTC technique, firstly, the input targets are encoded with an application of a phase mask, and then the joint power spectrum (JPS) is multiplied by an amplitude-modulated filter (AMF) of the reference object before the inverse Fourier transforming. An enhanced amplitude-modulated filter, in which the modulation factor varies according to the noise presented in the input scenes, is used in the PAJTC technique. The PAJTC technique eliminates extraneous signals especially for the JTC with multiple input targets. An optoelectronic implementation schematic diagram for the PAJTC technique and computer simulations of the performances of the classical JTC and the PAJTC for comparison are presented.
We present a possible way to detect 3D out-of-plane targets. Several Su-27 airplane images with different 3D rotational views were used to synthesize a template function, which successfully detects the target against the non-target such as the F16 airplane. A theoretical development for the purpose of pattern recognition is proposed. The system has the desirable property of sharp peaks with low sidelobes in the output correlation plane when multiple targets appear in the input. The test results show that the correlation peak is quite distinguishable at the location of the target and indicate the success of the technique. When combining the advantages of optics and electronics, the system is suitable for hybrid optical/electrical signal processing.
In many commercial applications in Chinese society, seals instead of signatures for person identification are widely used, such as money withdrawing lists, checks, receipts, etc. It needs high rate of correctness. However, it can not be effective if the process of safety validation is inspected merely by human eyes. Also, for safe control and management, in addition to the potential human cheating act, there is also a possible error due to human neglect. Based on the constrained energy minimization joint transform correlator (CEMJTC), we have proposed a prototype Chinese seal recognition system, which combines the advantage of light speed and programmability of computer. The CEM filter designed for optical correlator has the desirable property of sharp peak with low sidelobes levels in the output correlation plane. Th major advantages of the system are alignment simplicity, rotation invariant capability, high discrimination capability and suitability for hybrid optical signal processing.
We present a new technique for Chinese seal recognition using constrained energy minimization joint transform correlator (CEMJTC). Either optical or digital correlation is one of the most powerful operations for detecting the presence/absence of the seal image. The CEMJTC is proposed to improve the discrimination capability for shift- invariance and rotational-invariant seal recognition. By minimizing the average correlation energy with respect to all training seal images, while constraining the correlation peak value to a constant, a filter function is constructed. The main emphasis is to design a filter for good discriminate ability. Numerical results are presented to demonstrate the improvements. Furthermore, experimental results show the sharp correlation output profile when the seal image is correct, otherwise the correlation peak will be obviously reduced. The new technique for seal recognition shows a significant increase in high speed and detection ability.
Cell is the basic structural and fundamental unit of all organisms; the smallest structure capable of performing all the activities vital to life. One goal of current research interest is to learn how the muscle varies the strength of its contraction in response to electric stimuli. A wide variety of techniques have been developed to monitor the mechanical response of isolated cardiac myocytes. Some success has been reported either with the use of intact rat myocytes supported by suction micropipettes or in guinea pig myocytes adhering to glass beams. However, the usual measuring techniques exhibit destructive contact performance on live cells. They could not solve the problem, since the cell may die during or after the time-consuming attachment process at the beginning of each experiment. In contrast, a novel optical system, which consists of a microglass tube with an inner diameter the same size of a real cardiac cell, is proposed to simulate real cell's twitch process. the physical parameters of synthetic cell are well known. By comparing the dynamics of the real cell with that of the simulated cell, the twitch characteristics of the real cell can be measured.
Many correlation filters (e.g., matched spatial filters, phase-only filters, binary phase-only filters, etc.) are usually evaluated in terms of metrics such as signal-to-noise ratio, peak-to- correlation energy and Horner light efficiency. In this paper, we compare the MSF and POF using more direct performance measures, the probability of detection (PD) and the probability of false alarm (PFA).
In this work, a distortion-invariant pattern recognition scheme called the composite training image method is introduced. Usually, in attempting to detect the distorted (rotated, size- changed, shifted) versions of an object, a large number of raw training (distorted) images are used. However, there is a trade-off between this number and the ratio of signal correlation intensity peak to the maximum sidelobe (RSMS). In order not to degrade this ratio, the number of training images should be reduced as much as possible. We show how to fuse several similar raw training images into a composite training image. In this paper, we illustrate the feasibility of using such composite training images.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.