Grayscale Difference Research Articles

Research of FCM for image segmentation based on graph theory

Graph theory was utilized to improve image segmentation of traditional Fuzzy CMeans(FCM).The proposed algorithm used weighting of graph theory to calculate the distance of FCM,compared with Euclid distance,the proposed algorithm not only considered the distance of every sample,but also considered the Grayscale difference of every sample,and gained fuzzy membership function which was suitable for image segmentation.Based on the experimental result,probability of error and index of evaluation through comparing with image segmentation based traditional FCM and image segmentation based graph theory.The improved FCM in this paper is proved to be an appropriate method which is suitable for image segmentation.

Regional Analysis of Sinonasal Ciliary Beat Frequency

Mucociliary clearance, a primary host defense mechanism, depends on mucus production and its clearance by the coordinated beating of cilia lining the airways. Numerous investigations have analyzed ciliary activity in brushings from the inferior turbinate. To date, only one study has investigated whether there exists variation in ciliary beat frequency (CBF) within the sinonasal cavity. We analyzed CBF from the inferior turbinate, uncinate process, and sphenoethmoid recess in nonsinusitis patients to determine regional variability of ciliary activity within the sinonasal cavity. Explants of sinonasal epithelium were analyzed at 37 degrees C. Beating cilia were visualized with differential interference contrast microscopy. Images were captured using a high-speed digital camera with a sampling rate of 250 frames per second. A one-dimensional tracking algorithm analyzed individual pixel grayscale values within each frame of the video. The differences in grayscale were plotted as a time-dependent waveform, and frequency was calculated as the inverse of the peak-to-peak distance. A minimum of three areas of beating cilia were analyzed per regional sample. Statistical analysis was performed with repeated-measures analysis of variance. Complete sampling of all three sites was accomplished in 10 patients. No difference in CBF within the sinonasal cavity was identified (p < 0.05). The mean CBF for all sites in all patients was 12.6 +/- 2.9 Hz, in agreement with published values. This study shows no regional differences in CBF within the sinonasal cavity, supporting previous work and validating analysis of inferior turbinate cilia.

ディスプレイ―ＩＤＷ／ＡＤ’０５を中心に―シミュレーション画像を用いた主観評価に基づく新規な階調評価指標の提案

The patience limit (PL) and just noticeable difference (JND) of various grayscale viewing angle characteristics were evaluated subjectively using simulated images. Ten test subjects evaluated four types of picture images. Each type consisted of eight different images; two kinds of reverse images, three kinds of black crushing images, and three kinds of white omission images. All the images had several grayscale difference levels. The results showed that the concernments depend not only on the kind of pictures but also significantly on the images. We propose a new evaluation criterion, criterion grayscale difference odiousness (CGDO) value, for evaluating grayscale viewing angle characteristics. The value is calculated integrating the gamma-curve shape and the difference in the grayscale level from the original one. In our results, the CGDO values at PL and JND are about 2.0 and 0.5, respectively. Therefore the CGDO values shall be 2.0 or less, and should be 0.5 or less for ideal TFT-LCD legibility.

Biological gray scale for digital imagery

Biological evolution has adapted human vision to terrestrial light contrasts. Earthly scenes have a typical bimodal contrast histogram, with a reflection mode and a shadow mode. Consequently, human sensitivity to gray-scale differences also is bimodal. Luminance differences are most discriminable at the modal intensities of the terrestrial contrast distribution so that vision conveys maximum information about the world. By inverting this biologic, an image can be computationally adapted to the human visual system. Using Paul Whittle's model of gray-scale sensitivity as a basis, distinct pixel intensities in image data are mapped to optimally discriminable displayed luminances. Since this approach is scene dependent and display dependent, the optimum gray scale must be recomputed for each displayed image, background luminance and display environment.

New Perspectives for Visual Characterization of Pharmaceutical Solids

The utilization of descriptive image information in pharmaceutical powder technology is rather limited. Consequently, the development of this discipline is a challenge within physical characterization of pharmaceutical solids. The aim of this study was to develop and evaluate an inventive visual characterization approach for monitoring the granule growth in a fluidized‐bed granulation process and to use the generated image information in the prediction of tabletting behavior of granules. Surface images of samples from 34 granulations were continuously captured during the spraying and drying phases of the process and particle size distributions were determined. The gray scale difference matrix (GSDM) was derived from two surface images taken in controlled illumination conditions. The particle size calculation from the surface images was based on a multivariate Partial Least Square (PLS) model between the GSDM and sieve analysis measurements. The image information of the end‐point samples was also evaluated with respect to tabletting behavior of the granules produced. Principal component analysis (PCA) was used for data visualization. The introduced approach was suitable in particle size measurements of granules during all process phases and in the monitoring of different kinds of granule growth behavior. The visual inspection of the granule samples was powerful, enabling representational batch‐to‐batch comparisons. The tabletting behavior of the granules could be predicted directly from particle size information generated from the surface images. PCA as a projection method was efficient in data visualization. Development of process analytical technologies (PAT) aims at improving the efficiency of processes. The presented visual characterization approach can be an effective process analytical tool in particle size analysis also enabling the evaluation of the further product quality in the end of the granulation process. The idea of characterization of bulk surface images opens new perspectives for characterization of pharmaceutical solids. © 2004 Wiley-Liss, Inc. and the American Pharmacists Association J Pharm Sci 93:165–176, 2004

Analysis on imaging features of mammography in computer radiography and investigation on gray scale transform and energy subtraction

In this dissertation, a novel transform method based on human visual response features for gray scale mammographic imaging in computer radiography (CR) is presented. The parameters for imaging quality on CR imaging for mammography were investigated experimentally. In addition, methods for image energy subtraction and a novel method of image registration for mammography of CR imaging are presented. Because the images are viewed and investigated by humans, the method of displaying differences in gray scale images is more convenient if the gray scale differences are displayed in a manner commensurate with human visual response principles. Through transformation of image gray scale with this method, the contrast of the image will be enhanced and the capability for humans to extract the useful information from the image will be increased. Tumors and microcalcifications are displayed in a form for humans to view more simply after transforming the image. The method is theoretically and experimentally investigated. Through measurement of the parameters of a geometrically blurred image, MTF, DQE, and ROC on CR imaging, and also comparison with the imaging quality of screen–film systems, the results indicate that CR imaging qualities in DQE and ROC are better than those of screen–film systems. In geometric blur of the image and MTF, the differences in image quality between CR and the screen–film system are very small. The results suggest that the CR system can replace the screen–film system for mammography imaging. In addition, the results show that the optimal imaging energy for CR mammography is about 24 kV. This condition indicates that the imaging energy of the CR system is lower than that of the screen–film system and, therefore, the x‐ray dose to the patient for mammography with the CR system is lower than that with the screen–film system. Based on the difference of penetrability of x ray with different wavelength, and the fact that the part of the x‐ray beam will pass through the image plate in the procedure of CR imaging, the method of subtraction of the two images which were taken in the same time with one exposure can increase the diagnostic information. Image registration for mammography with CR imaging is usually ignored because the two images are taken in one exposure time. This dissertation investigated the necessity of image registration for image energy subtraction in CR mammography imaging. A novel method for image registration that can reduce the computing time is established, based on the features of CR imaging for mammography.

Characterization of particle sizes in bulk pharmaceutical solids using digital image information.

The purpose of this study was to demonstrate a novel method of extracting relevant information from undispersed bulk powder surfaces to be used in particle size analysis. A new surface imaging approach for undispersed powders combined with multivariate modeling was used. Digital surface images of various granule batches were captured using an inventive optical setup in controlled illumination conditions. A descriptor, the gray scale difference matrix (GSDM), which describes the particle size of granular material was generated and extracted from the powder surface image information. Partial least squares (PLS) modeling was used to create a model between the GSDM and the particle size distribution of granules measured with sieving. The use of lateral illumination and the combining of information from 2 surface images strengthened the shading effects on the powder surfaces. The shading effects exposed the topography or the visual texture of the powder surfaces. This textural information was efficiently extracted using the GSDM descriptor. The goodness-of-fit (R2) for the created PLS model was 0.91 and the predicted variation (Q2) was 0.87, indicating a good model. The model covered granule sizes in the size range of approximately 20 to 2500 microm. The extracted descriptor was effectively used in particle size measurement. This study confirms that digital images taken from undispersed bulk powder surfaces contain substantial information needed for particle size distribution analysis. The use of the GSDM enabled the utilization of bulk powder surface information and provided a fast method for particle size measurement.

Solution of pattern matching inspection problem for grainy metal layers

In this paper, we demonstrate a new method of improving the defect controllability of grainy metal layers, for example, Hot-Al-Cu wiring, by enhancing the practical sensitivity of in-line inspectors. The problem in increasing practical sensitivity is the nuisance counts caused by grain boundaries, which do not cause electrical failures. We propose a method of decreasing the signal from the grain boundaries. On the grayscale images taken by pattern matching inspectors, grain boundaries are observed as gray on Hot-Al-Cu wiring, which is observed as white. If the illumination brightness is increased, the gray level of the grain boundaries becomes higher and saturates at the upper limit of grayscale, i.e., white. On the other hand, the gray level of the wiring stays white. Thus the signal, the grayscale difference between the grain boundaires and the wiring, can be decreased to almost zero. We call this phenomenon the effect. Our experimental results prove that the saturation effect due to illumination brightness optimization successfully cancels the nuisance counts caused by grain boundaries. The practical sensitivity limit is enhanced from 0.8 to 0.4 /spl mu/m. This solution will greatly improve the defect controllability of grainy metal layers.

Soft-Copy Quality Control of Digital Spot Images Obtained by Using X-ray Image Intensifiers

To evaluate 12 x-ray image intensifier (XRII) digital spot systems. Four objective tests were performed to assess XRII digital spots: entrance exposure, patient exposure, soft-copy gray scale, and pixel noise. Two additional tests were performed to assess high-contrast limiting resolution and threshold contrast detection. Digital spot XRII entrance exposures averaged 1 x 10(-7) C/kg (0.38 mR) for units with large fields of view (FOVs); mean entrance exposure in a medium-sized patient was 1. 25 x 10(-5) C/kg (48 mR). Luminance measurements of the table-side monitors provided a mean of 473 just-noticeable differences in gray scale with the room lights off. Mean resolution with a bar test pattern was measured as 1.5 line pairs per millimeter for systems with a 40-cm FOV. Measured pixel noise (in relative units) was 6-25. Mean threshold contrast with the lights off was 0.85%. Once input exposure is normalized for FOV and image matrix size, soft-copy assessment of limiting resolution with either low-contrast detection or, preferably, an off-line noise metric (pixel SD) provides objective measurements of digital spot image quality. With the lights on, 10 systems with room-light sensors had an 11% loss of gray scale. For systems without sensors, the loss was 33%.

Cluster update algorithm and recognition

We present a fast and robust cluster update algorithm that is especially efficient in implementing the task of image segmentation using the method of superparamagnetic clustering. We apply it to a Potts model with spin interactions that are are defined by gray-scale differences within the image. Motivated by biological systems, we introduce the concept of neural inhibition to the Potts model realization of the segmentation problem. Including the inhibition term in the Hamiltonian results in enhanced contrast and thereby significantly improves segmentation quality. As a second benefit we can, after equilibration, directly identify the image segments as the clusters formed by the clustering algorithm. To construct a subsequent spin configuration the algorithm performs the standard steps of (i) forming clusters and of (ii) updating the spins in a cluster simultaneously. As opposed to standard algorithms, however, we share the interaction energy between the two steps. Thus, the update probabilities are not independent of the interaction energies. As a consequence, we observe an acceleration of the relaxation by a factor of 10 compared to the Swendson and Wang [Phys. Rev. Lett. 58, 86 (1987)] procedure.

Magnetization reversal of a thin polycrystalline cobalt film measured by the magneto-optic Kerr effect (MOKE) technique and field-dependent magnetic force microscopy

The magnetic properties of a 100-nm-thick cobalt film, deposited thermally on a glass substrate under high-vacuum conditions, were measured ex situ through the longitudinal magneto-optic Kerr effect (MOKE) technique and through field-dependent magnetic force microscopy (MFM). The MOKE measurements showed that the film had a uniaxial anisotropy in the plane of the sample. Two series of MFM images were measured under the presence of an external magnetic field, applied separately along the easy and hard axes, with magnitudes encompassing positive and negative saturation of the sample magnetization. Each series of images recorded the magnetization reversal clearly, revealing and explaining distinctions between the two reversal processes that correlate well with the macroscopic reversals recorded by the MOKE technique. The reversal of the gray-scale polarity with the change of the applied field from negative saturation to positive saturation suggests the presence of a component to the magnetic force that was related directly to the magnetization of the individual grains. A calculation of the easy-axis MFM gray-scale image differences showed a hysteresis-like behavior that agreed qualitatively with the corresponding measured MOKE hysteresis loops.

Population Studies of Titanium-Bearing Inclusions in AISI 316l using Automated Image Analysis with Scanning Electron Microscopy/Electron Dispersion Spectroscopy (SEM/EDS)

Abstract Titanium is used in the manufacture of AISI 316L austenitic stainless steel as a deoxidizer. Titanium is considered a residual element, and limits are often not specified. As a residual element, titanium most often appears in complex oxide inclusions, usually in conjunction with aluminum and other deoxidizers. Titanium-bearing inclusions have been found to initiate pitting during electropolishing. The present work was performed to evaluate the acceptable upper limit for titanium as a residual element in AISI 316L. Metallographic specimens were prepared from seven commercial heats of AISI 316L (Table 1) with titanium contents varying from 0.003 % to 0.014 %. A quantitative metallographic analysis of each specimen was performed using an SEM/EDS system with an automated stage. Operating parameters are given in Table 2. Data was obtained with the system in the backscatter (BS) mode, producing a high-contrast image based on differences in chemical composition. Once each inclusion was identified by its difference in gray scale, it was also analyzed by EDS for elements present.

Chromatic Response Curves of Surface Colours

Basic data in colour vision such as chromatic response curves were obtained with aperture colours. It is not possible to apply the cancellation technique to define redness or yellowness components etc in a surface colour. A method was developed to obtain such information (Indow, 1987 Die Farbe34 253 – 260), which was based on the expanded form of multidimensional scaling. The results by an improved method are presented. First, it is shown how multiattribute perceptual differences between two Munsell colours, P j, P k, are predicted by distances ^ djk in the current Munsell solid. Colour differences djk are defined by matching with lightness differences in Munsell gray scale and root-mean-squares of ( djk - ^ djk) is 0.2∼0.3 in the unit of Munsell Chroma C. Second, principal hue vectors falpha, alpha= R, Y, G, B, or P in addition, are defined in the current Munsell solid. As shown before, fB is shifted from 5 R in the direction of 5 P, irrespective of whether P is included or not. Overall root-mean-squares of (scaled values of principal hue a in a colour j-coordinates of P j on falpha) is about 0.6 in the unit of C. On the basis of these findings, it becomes possible to define principal hue components for any Munsell colours and hence chromatic response curves are obtained as a function of H, V, and C.

A desktop image processing and photogrammetric method for rapid volcanic hazard mapping: application to air-photo interpretation of Mount Pelée, Martinique

A desktop image processing and photogrammetric method was developed for digitizing black-and-white aerial photographs. The technique was applied to airborne optical images of Mt. Pelee, Martinique, a historically active volcano in the tropical Lesser Antilles island arc, to evaluate its utility for rapid geologic mapping and hazard assessment in vegetated areas. The digital approach provides several advantages over traditional air-photo interpretation by allowing for change detection in time-series images, morphologic characterization, development of digital elevation models from stereopairs, and geo-referencing with other digital data sets. A digital mosaic of Mt. Pelee was created from air photos acquired in 1951, which covered the region affected by the 1902 eruption. Severe mismatches occurred along edges of adjacent photographs prior to correction, which precluded quantitative morphologic analysis of the volcanic edifice. Geometric corrections and histogram equalization of digitized air photos allowed creation of a continuous mosaic. Comparison of the mosaic and a map based on differences in gray scale and texture to a volcanostratigraphic map revealed that not only the various deposits produced during the 1902 event were easily differentiated, but that older eruptive products were identified, suggesting that this approach may be used for rapid hazard evaluation of historically active tropical volcanoes.

Motion estimation via cluster matching

A new method for estimating displacements in computer imagery through cluster matching is presented. Without reliance on any object model, the algorithm clusters two successive frames of an image sequence based on position and intensity. After clustering, displacement estimates are obtained by matching the cluster centers between the two frames using cluster features such as position, intensity, shape and average gray-scale difference. The performance of the algorithm was compared to that of a gradient method and a block matching method. The cluster matching approach showed the best performance over a broad range of motion, illumination change and object deformation. >

Gray scale and achromatic color difference

Gray scale is a preferred method for information display, yet existing techniques for constructing a gray scale rely on mathematical formulas that are ambiguous and inconsistent with known phenomena of suprathreshold achromatic perception. To improve this situation, problems are discussed, literature on suprathreshold achromatic perception is reviewed, existing gray-scale formulas are compared with human-performance data, and a new approach is examined. The new approach takes into account the luminances of the surround, the targets, and the distractors that are to be rendered in a gray scale. It is theory based, is consistent with known phenomena of suprathreshold achromatic perception, and fits human-performance data at least as well as any other formula.

1.イメージサイズのちがいと画質について : 同一情報密度における信号認識限界で比較(I.RI画像構成とその評価)(第37回総会)

1.イメージサイズのちがいと画質について : 同一情報密度における信号認識限界で比較(I.RI画像構成とその評価)(第37回総会)

Initial observations with the SOLWIND coronagraph

Since 1979 March 28 a new NRL coronagraph called ''Solwind'' has been observing the Sun's outer corona (2.6--10.0 R/sub sun/) routinely with a spatial resolution of 1'.25 and a temporal resolution of 10 minutes. These are the first satellite observations of the outer corona near the peak of a sunspot cycle, when coronal transients and high-latitude streamers are expected to be common. Coronal variations on time scales from 10 minutes to 27 days are revealed dramatically in easily constructed gray-scale difference images.