Search

The study of image evaluation is a complex subject requiring the simultaneous consideration of the object, optical system, detector, and mechanism for processing the detector's signals. Image quality for any given photo-optical system depends upon such diverse considerations as granularity, light scattering, numerical aperture, tone reproduction, the degree of correction in the image-forming optics, etc.

A number of signal processing techniques which have been developed for processing one-dimensional sequences do not generalize to the processing of two-dimensional signals, largely due to the absence of a two-dimensional factorization theorem. In an attempt to circumvent this problem, a specific representation of two-dimensional sequences as one-dimensional sequences is presented in this paper. Using this mapping several two-dimensional problems can be viewed as one-dimensional problems and approached using one-dimensional techniques. This representation is valid both for signals of finite extent and for the more general class of signals with rational Z-transforms. In this paper we consider applications of these techniques for high speed convolution, processing of drum scans, and two-dimensional finite impulse response (FIR) filter design.

This paper describes signal processing in the encrypteddomain, i.e., that after encryption but before decryption. In this framework, signal processing operations can be directly applied to encrypted signals without decrypting of encrypted signals, whereas the ordinary framework encrypts signals for transmission and/or storing but it decrypts them before signal processing operations are applied to. The described framework befits contemporary cloud computing in which not only transmission but also storing and processing are done in the public Internet. Addition to brief survey, two tangible application scenarios are also demonstrated in this paper where a new signal processing algorithm is introduced each.

A feasible pavement crack detection system plays an important role in evaluating the road condition and providing the necessary road maintenance. In this paper, a back propagation neural network (BPNN) is used to recognize pavement cracks from images. To improve the recognition accuracy of the BPNN, a complete framework of image processing is proposed including image preprocessing and crack information extraction. In this framework, the redundant image information is reduced as much as possible and two sets of feature parameters are constructed to classify the crack images. Then a BPNN is adopted to distinguish pavement images between linear and alligator cracks to acquire high recognition accuracy. Besides, the linear cracks can be further classified into transversal and longitudinal cracks according to the direction angle. Finally, the proposed method is evaluated on the data of 400 pavement images obtained by the Automatic Road Analyzer (ARAN) in Northern China and the results show that the proposed method seems to be a powerful tool for pavement crack recognition. The rates of correct classification for alligator, transversal and longitudinal cracks are 97.5%, 100% and 88.0%, respectively. Compared to some previous studies, the method proposed in this paper is effective for all three kinds of cracks and the results are also acceptable for engineering application.

The purpose of this paper is to describe the system for the display, processing, and production of image-data products created to support the Mariner 9 Television Experiment. Of necessity, the system was large in order to respond to the needs of a large team of scientists with a broad scope of experimental objectives. The desire to generate processed data products as rapidly as possible, coupled with the complexities introduced by the nature of the vidicon camera, greatly increased the scale of the ground-image processing effort. This paper describes the systems that carried out the processes and delivered the products necessary for real-time and near-real-time analyses. References are made to the computer algorithms used for the different levels of decalibration and analysis.

This paper presents an innovative method of face detection by supplementing dynamic time warping algorithm with proposed image processing strategy and weighting scheme. Using our proposed approach overcomes some of the shortcomings in applying dynamic time warping for face detection, hence improving the performance and detection accuracy. The results presented and discussed in this paper show the efficacy of our approach in using DTW for face detection. Ill. 16, bibl. 16, tabl. 1 (in English; abstracts in English and Lithuanian).http://dx.doi.org/10.5755/j01.eee.108.2.154

This paper is the result of editorial license in the effect that it is designed to augment the invited papers for enhancement and restoration in this issue. Unfortunately this editor found the genesis of a special issue to be a very dynamic process, and as such this paper has experienced many contortions. However, it is hoped that a few topics not covered earlier, but which warrant some discussion in a digital image enhancement issue, can find a home (albeit brief and uneasy) in this editorial. The topics discussed below include some heuristic techniques which might be useful to the image enhancement objective, some superresolution and positive restoration techniques (still of esoteric interest), some comments on experimental methods, and a bibliography restricted to digital techniques for image restoration and/or enhancement.

In these recent years, Compressive Sensing (CS) is becoming an attractive topic in the field of Information Theory. It is widely used in several area including networking, image processing and digital camera. In particular, image reconstruction based on small number of measured components is known as the most useful application. In this paper, SL0 algorithm is specially used for the reconstruction process. It significantly decrease the processing time by utilizing a matrix in which the number of row is much smaller than number of column. Therefore, SL0 is known as one of the fastest and most accurate algorithm in CS. However due to ill-posed condition, if the prior information of the original image is undetermined, the reconstruction procedure of SL0 is much affected by the noise. Unfortunately, the investigation for solving this SL0 ill-posed condition is very limited therefore SL0 is not widely applied in many application. Consequently, this paper proposes a novel regularization technique for SL0 algorithm in the frequency domain. In order to reduce and constraint the space of reconstructed image, the frequency domain Tikhonov regularization technique is employed. It is shown that the quality of the reconstructed image is much better compared to the traditional algorithm under the noisy environment. The experimental result is exclusively simulated for 3 images: Lena, Sussie and Cameraman under both Gaussian and Non-Gaussian noise models (such as AWGN, Poisson noise, Salt & Pepper noise and Speckle noise) at different noise powers.

Topographic mapping can be considered as the combination of processes required to produce a topographic map. Coherent optics has the potential for playing many roles in mapping systems of the future. The purpose of this paper is to indicate some of these roles. The limited discussion must be considered as an initial ad hoc attempt to define potentials in the eight areas of Photogrammetric Data Reduction, Image Processing, Optical Memories, Pattern Recognition, Performance Prediction, Holographic Terrain Displays, All-weather Mapping, and Field Applications. Other papers from the proceedings are broadly integrated along with selected references.

In the process of real time image processing, the amount of image processing computing and data is very large, the use of single core or a single DSP can not meet the needs of real-time image processing. This paper designs the new generation multi-core DSP real-time image processing system, the system is a multi-core high performance DSP as the core circuit, and also uses FPGA counterweight device, high-speed data interface and high-speed external memory, to improve the data cache speed and data bandwidth, and the system satisfies the design of low loss and high computational performance. In order to verify the validity and reliability of the system, this paper uses the Hyper Lynx software to carry on the simulation test for the system, and sets the IP and sub net mask of image real time processing to obtain the eye and timing diagram of real-time image processing. Finally, the system image processing performance results can be obtained by the joint debugging, the simulation test results can be seen that the speed of real-time image processing is quick, and the rate of correct recognition is high, which can satisfy the need of real-time image processing, to provide the technical reference for the design of real time image processing system

Xeroradiography is the technology for recording radiographic images without the need for any chemical processing whatsoever. To make an image a selenium photoreceptor, or plate, is charged to a uniform surface potential on the order of 1000 volts. When this photoreceptor is exposed to x-rays, its surface charge is depleted by photoconduction in amounts proportional to the radiation intensity striking it. A residual potential pattern which is a shadow graph of the object being x-rayed is then formed. After exposure, the electrostatic image on the plate is developed by a powder cloud technique in which charged, pigmented particles are blown over the plate and adhere to it in the pattern of the image. This powder or toner is the ink which defines the image. Following development, the toner is transferred to paper and fixed into a permanent image by low temperature heat fusing. The purpose of this paper is to present the principles of zero radiography, to describe practical equipment utilizing these principles for x-ray imaging in the medical and industrial fields, and to show some examples of the types of images produced by the system.

Two years ago at the tenth SPIE Symposium I presented a paper (Ref. 1) which illustrated some of the work JPL has been doing in the development of hardware and techniques for processing of the Ranger and Mariner pictures. It is the purpose of this paper to bring you up to date on our more recent activities.© (1967) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

This paper deals with a program of research in which the techniques and fundamental limitations of the restoration of degraded images is receiving study. The research program includes all forms of image degradation but this paper is concerned only with the case of degradation of the type encountered in looking up through the atmosphere with a large ground based telescope.© (1967) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

This paper treats the problem of restoring the detail to an optical image which has been degraded by diffraction and noise. The particular contribution of the paper is a more complete analysis of the effects of various types of noise on system performance than has been given previously. Background noise, measurement noise, and computer roundoff error are considered, and the errors in the reconstructed image caused by these noise processes are evaluated. Numerical results for the special case of a perfect one-dimensional slit aperture are obtained. A general conclusion is that the reconstruction technique described here is most useful when the smoothing is severe and when a modest improvement of resolution may be worthwhile.

This paper describes a process which produces finished poisitive pictures, directly from the camera, in about one minute after the exposure. The camera is described as containing a strip of paper in addition to the negative material and as feeding this strip of paper, in contact with the exposed negative, through simple pressure rollers and thence out of the camera box. The process to be used with the camera is described as having but one step, instead of the many steps of conventional photography; and it is shown that this can be achieved by associating with the strips a reagent which, when spread between the two strips by the pressure of the rollers, is capable of developing the silver halide of the negative and forming the positive image at one and the same time. Several classes of processes are discussed, their characteristics explored and certain principles established for obtaining satisfactory picture quality, stability, speed of operation, etc. In one of these processes, the reagent spread between the negative and the positive strip consists of a small amount of viscous liquid containing, in rather high concentration, the necessary constituents for developing the negative image, for forming at the same time a silver complex with the unexposed grains in the negative, for transferring the soluble complex to the positive sheet, and there creating and stabilizing the positive silver image. This process runs to completion in about one minute. When the two strips are peeled apart, both are essentially dry. One strip is the finished positive picture. The process operates at temperatures from less than 30°F to over 100°F. The paper discusses the control of the rates of reduction, silver ion complex formation, and ion diffusion in the several reaction fronts; principles of stabilization of the positive picture; control of the relative rates of growth of density in negative and positive; conversion of the silver ions to particles of silver of adequate size and the control of the color of the image as a function of particle size; and various characteristic curves which have been obtained with different negative materials. These factors are then interpreted with relation to the photographic usefulness of the process.

Spatial frequency response techniques have been used in the past for the evaluation and design of image processing systems. In this paper, these techniques are extended to permit a statistical analysis of degradation in scanned image systems. The analytical techniques used have been developed by engineers for the statistical analysis of sampled data systems. The use of these techniques allows an evaluation of the average degradation of the actual images to be processed, rather than the degradation of hypothetical test images. In addition, the statistical figure of merit used to measure degradation appears to be a good measure of human readability of degraded images. General expressions for degradation in terms of scan density, the system transfer functions, and statistical parameters of the images are derived in this paper for several classes of linear scanned image systems.

The spatial and spectral luminance properties of a cloud's surface are being studied to gather more information about the microphysical processes occurring within the cloud. A real-time, field-implementable image processing system has been developed by the authors to sense and display this luminance information, and is described in this paper. Applications, including the investigation of “crown flash” phenomena, are described.

This paper describes the fabrication and demonstration of a real-time incoherent-to-coherent optical converter having applications in image processing systems. Briefly, it utilizes the photoconductivity effect in the single-crystalline electrooptic material, Bi 12 SiO 20 to spatially modulate its electrical polarization. An optically absorbed write-in image is stored as an image wise polarization pattern in the device. Readout is accomplished electrooptically by means of the subsequent phase retardation that a polarized beam of coherent light undergoes in passing through the Bi 12 SiO 20 . An operating mode for achieving continuous image conversion with high-speed recyclability is described. The following performance characteristics have been demonstrated : write-in intensity of 300 µW/cm2at 400 nm yielded a contrast ratio of 50:1 after 40-ms exposure. When the converter was operated at a frame rate of 10/s, a sampled read-out contrast ratio of greater than 1000:1 was achieved. Resolution in excess of 80 1p/mm has been recorded and read out. The crystal growing and device fabrication methods by which 1-square-in converters have been built and operated are described. Results achieved in using this device to Fourier transform images are also presented.

Rat liver histological sections were photographed through a Leitz microscope and the negative digitised using a P-1000 Photoscan micro-densitometer (OPTRONICS). In the system the transmitted light through the film was measured using a photo-detector, converted to 256 grey levels, and stored on a 7 track magnetic tape. The optical density was measured every 100 microns. The final picture consisted of 750,000 points. These were fed into a PDP-15/40 computer with core requirements of 24 K words and two 250 K word disks, and analysed. The present paper describes a detailed analysis of one heptocyte.

The application of digital image processing techniques to medical problems has been under development at the Jet Propulsion Laboratory, California Institute of Technology, since 1966. The initial research concentrated on medical X-ray films and include (1) image restoration, i.e., using digital techniques to retrieve information partially lost in the imaging system which generated the pictures, (2) image enhancement to emphasize features or remove useless data, and (3) image mensuration to provide more quantitative measurements and analyses than are available through visual inspection techniques. From the initial concern with X-ray image processing, this research in image processing as applied to biomedicine has been expanded to include a system for the automated analysis of images from light and electron microscopes. The Automated Light Microscope System (ALMS), around which the chromosome analysis system has been developed, allows direct scanning of specimen images into a computer for subsequent analysis. The applications of digital image processing to biomedicine at the Jet Propulsion Laboratory encompass 15 separate research projects. The areas selected for discussion in this paper give some indication of the types of picture analysis that are being performed. The current trend in biomedical image processing is toward more human interaction in the research setting.

This paper describes an algorithm for the approximate solution of vector Toeplitz systems via an iterative-improvement method. The algorithm exploits the special structure of Toeplitz matrices, namely, their similarity to vector circulants, and is particularly well suited for solving large systems. Sufficient convergence conditions and concrete error bounds for the iteration are presented along with an application of the routine to a problem in the design of planar digital filters for image processing.

Recursive restoration of two-dimensional noisy images gives dimensionality problems leading to large storage and computation time requirements on a digital computer. This paper shows a two-dimensional second-order Markov process representation can be used for fast recursive restoration of images with small storage requirements. Advantages of this method over existing techniques are illustrated by means of examples.

This paper describes various forms of line drawing representation, compares different schemes of quantization, and reviews the manner in which a line drawing can be extracted from a tracing or a photographic image. The subjective aspects of a line drawing are examined. Different encoding schemes are compared, with emphasis on the so-called chain code which is convenient for highly irregular line drawings. The properties of chain-coded line drawings are derived, and algorithms are developed for analyzing line drawings to determine various geometric features. Procedures are described for rotating, expanding, and smoothing line structures, and for establishing the degree of similarity between two contours by a correlation technique. Three applications are described in detail: automatic assembly of jigsaw puzzles, map matching, and optimum two-dimensional template layout

In the formation of an optical image, each surface element of the object gives rise to a more or less blurred distribution in the image surface, of total brightness proportional to that of the object element. The image is the sum of these distributions in the appropriate sense: when the object is coherently lit, the image is built up by adding their complex amplitudes; when the object elements are regarded as incoherent it is the intensities which are added. In both cases the image can be expressed as the convolution of the object with a spread function which characterizes the optical system. In systems for which the spread function does not change appreciably from one part of the field to another, the Fourier transform of the image is obtained to a sufficient approximation on multiplying the Fourier transform of the object with that of the spread function. More generally, this holds for any part of the field of a non-isoplanatic system over which the changes in the form of the spread function are small enough to be disregarded; we call such an area an 'isoplanatism-patch'. Working over such an area, an optical system can be regarded as a linear filter in which the Fourier components of the object reappear in the image multiplied by 'transmission factors'. These factors, first considered by Duffieux, depend on the aperture and aberrations of the system, and in Section 2 they are evaluated in terms of an ikonal function. The qualities required of an optical image are so varied that an assessment valid over the whole range of practical applications seems out of the question. Two extreme cases are considered in the present paper. In the first of these it is assumed that the aim of an optical design is to produce an image which is directly similar to the object. This is appropriate when no process of image interpretation or reconstruction is envisaged. In the second case, the aim is to produce an image containing the greatest possible amount of information about the object, without regard to the complexity of the interpretation processes which may be needed to extract it. For the first case, a criterion of image fidelity is proposed in Section 2.4 which gives a numerical measure of the resemblance of image to object in terms of the transmission factors of the optical system. In the second case, assessment is based on the information content of the image in Shannon's sense. This depends not only on the transmission factors of the system but also on the statistical properties of the presumed object set and of the unpredictable fluctuations which necessarily disturb observation; the analysis is carried through in Section 3. In Section 4 the assessment of optical images is discussed in terms of these two criteria.

In this paper we present some new ideas on pattern recognition of bubble chamber photographs. We have found global transformations that reduce circular tracks to a point in a two-dimensional angle-curvature space. The performance of this tracking method on a sample of bubble chamber events is presented and planned improvements are discussed.

The approximation of arbitrary two-dimensional curves by polygons is an importanttechnique in image processing. For many applications, the apparent ideal procedure is to represent lines and boundaries by means of polygons with minimum number of vertices and satisfying a given fit criterion. In this paper, an approximation algorithm is presented which uses an iterative method to produce polygons with a small—but not minimum—number of vertices that lie on the given curve. The maximum distance of the curve from the approximating polygon is chosen as the fit criterion. The results obtained justify the abandonment of the minimum-vertices criterion which is computationally much more expensive.

Impulse noise removal is a very important preprocessing operation in many computer vision applications. Usually it is accomplished by median filter with excessive sorting and therefore large power. This paper presents a new design of 2D median filter that utilizes a simple conditional filtering technique, executes fewer computations than related designs while achieving superior image quality. Experimental FPGA implementation of the proposed filtering scheme is compact, fast and low-power consuming.

Hough has proposed an interesting and computationally efficient procedure for detecting lines in pictures. This paper points out that the use of angle-radius rather than slope-intercept parameters simplifies the computation further. It also shows how the method can be used for more general curve fitting, and gives alternative interpretations that explain the source of its efficiency.

Walsh functions have become quite useful in the applications of image processing and feature selection. Due to their inherent efficiency of implementation, (they are a subset of the Reed Muller Codes and Hadamard matrices), they have become popular for coding, enhancement and other signal processing tasks. This paper will briefly describe applications in some of these areas with emphasis on a correlation analysis for justification purposes.

In this paper we present a machine vision system for the automatic inspection of defects in textured surfaces found in industry. The defects to be inspected are those that appear as local anomalies embedded in a homogeneous texture. The proposed method is based on a Gabor filtering scheme that computes the output response of energy from the convolution of a textured image with a specific Gabor filter. The best parameters of a Gabor filter are selected so that the energy of the homogeneous texture is zero, and any unpredictable defects will generate significantly large energy values. A simple thresholding scheme then follows to discriminate between homogeneous regions and defective regions in the filtered image. This transforms texture differences into detectable filter output. The experiments on structural textures such as textile fabrics and milled surfaces, and statistical textures such as leather and ,sandpaper have shown the effectiveness of the proposed method.

A data processing system which transforms photographic scans of the moon into rectified lunar pictures has been developed during the LUNAR AIMP satellite program of the National Aeronautics and Space Administration's Goddard Space Flight Center. Three coplanar narrowbeam optical sensors mounted on the spinning satellite provide the photographic input to the system. The satellite's processor determines what lunar image data to transmit, delta-modulating and interleaving these data with necessary navigational parameters. The ground processor reconstitutes the delta-modulated image data into lunar pictures, arranged so as to give the appearance of being the projection of the lunar surface onto one of 24 projection planes. This conversion is performed by first locating each lunar element corresponding to an item of image data. Next the projection plane is selected, either as input or by the program. The lunar elements are then mathematically projected onto the plane. This projection is formatted onto magnetic tape, from which the output picture is generated. In the paper, particular attention is paid to the information flow in the overall spacecraft/ ground-processor system. Some reisults of simulations of portions of the systems by personnel of NASA, Goddard Space Flight Center, and the Image Processing Department of the Federal Systems Division of International Business Machines Corporation (IBM), and of the processing of moon surface imagery on IBM's Electro-Optical Simulation Facility (EOSF) are provided.

This paper presents an automated scoring system for the twist skill in Horizontal bar based on motion image analysis. In this system, training scenes of Horizontal bar are taken by a high-speed video camera, and then a gymnast’s region is extracted from a video image frame by frame based on a background subtraction method. Next, the body axis of the gymnast in the moment when a twist skill completed is estimated from the gymnast’s region. Finally, the deduction point for the twist skill is automatically decided according to the rules described in the Code of Points. In experiments using video images for 26 practices, it was shown that about 80.8% of the correspondence rate between estimated deduction scores and true ones calculated by the correct ‘SCF’ and ‘Body axis’ was obtained. This can be promising as a result at the preliminary stage of this research.