Search

We report the influence of the rubbing treatment of the alignment layers on the electro-optic characteristics of AFLCD's. We applied a different rubbing treatment on different regions within a pixel. This variation of rubbing strengths causes domains with different surface interaction, which results in a different threshold voltage for each domain. In-pixel variation of the rubbing treatment and of the surface interaction influences the double hysteresis of the transmission-voltage characteristic, which is made less steep. The difference between the voltages at which the transmission starts to increase and the voltage at which the transmission reaches its maximum increases and so the selection of analogue grey levels is made easier. The results also show that the finite slope of the double hysteresis which is observed in normal AFLCD's, without extra rubbing treatment is probably caused by variations of the surface interaction.

Flotation processes occurring in the bulk and froth phases have a characteristic influence on the structural features and dynamics of the flotation froth. In principle the froth features can therefore be used as a useful indicator of the performance of the flotation system. In this study the surface froth features and dynamics are represented by three features extracted from the digitized images of the froths, viz. a statistical feature which is a rough indication of the average bubble size of the froth, a measure of the froth stability, as well as the average grey level of the froth, which is an indication of mineral loading. The effect of high intensity conditioning on the batch flotation of a sulphide ore from the Merensky reef in South Africa was investigated, and the significantly beneficial effect of high intensity conditioning on the performance of the flotation was clearly reflected in the smaller bubble size distributions and greater stability of the flotation froths.

Presenting 3D images produced by confocal optical slicing techniques in a static and easily publishable form can be difficult. Here, we demonstrate the presentation of the data as a single image random dot stereogram (SIRD), which can be viewed as a 3D object by ‘defocusing’ the eyes. The production of the SIRD employs three steps: (i) acquisition of the optical slices using confocal techniques, (ii) allocation of a suitable grey level to the object in each slice to provide depth-encoding information for the final image and (iii) calculation of the SIRD from the composite depth-encoded image. The technique is demonstrated with a limited number of optical slices through an acridine-orange-stained neutrophil (diameter =10 μm), in order to show the relative positions of the nuclear lobes in the cell.

Image analysis methods were developed for light and epifluorescence microscopy of Streptomyces ambofaciens undergoing differentiation in submerged culture. Grey level images were obtained with an integration controlled CCD camera and allowed three parameters to be measured: occurence of empty zones in mycelium, number of septations, mycelium thickness.

The application of digital image processing to mass screening for breast cancer is demonstrated in the context of adaptive spatial filtering. Experimental results for feature extraction by texture analysis and grey level thresholding reveal there is significant potential for computer-aided mass screening as an adjunct to radiological examination. Digital image processing can improve the visibility of image features, such as neoplasms or tumours, providing also speed and automation of inspection, whilst saving the radiologist time which is otherwise spent on checking clear samples (which represent the majority of screening cases). Zusammenfassung. Die Anwendung ziffernmasiger Bildbearbeitung zur Massenabblendung fur Brustkrebs wird im Zusammenhang mit raumlichem Filtern demonstriert. Experimentresultate fur Besonderheiten-Extraktion im Falle von Gewebeanalyse und Grau-Niveaubegrenzung enthullen, das hier eine bedeutende Moglichkeit fur computer-unterstutzte Massenabblendung als ein Zusatz zur radiologischen Untersuchung besteht. Ziffernmasige Bildbearbeitung kann die Sichtbarkeit von Bildmerkmalen, wie Neoplasmen und Tumoren, verbessern, auserdem Geschwindigkeit und Automation der Inspektion bietend, wahrend dem Radiologen Zeit ersparend, die andernfalls zur Uberprufung klarer Muster verwendet wurde (was fur die Mehrzahl der Abschirmfalle zutrifft).

Road traffic accidents are a problem which is countered by the development of systems that can minimize the number of fatal accidents by providing warnings to the driver, in particularly by vision-based driver assistance systems (VBDAS). Vehicle detection at night-time is very complex compared to day time due to availability of limited features and different illumination conditions. When driving at night-time, vehicles approaching from front are only visible by their headlights. This paper presents a monocular vision system capable of detecting vehicles in front views using a Haar-like feature approach in night-time gray-level video sequences. The approach detects vehicles at night-time using a camera by searching for headlights. Experiments demonstrate the effectiveness of the proposed system.

In recent years, a lot of 2D textures have been synthesized from input 2D textures. However, the quality problems still exist for many textures. Further improvements are required to extract more reliable texture features. In this paper, we present a texture synthesis approach using cooperative color and grey-level features. For color feature extraction, we extract appearance vectors to replace RGB color values. For grey-level feature extraction, we extract the statistical features including entropy, contrast, and correlation based on the grey level co-occurrence probabilities (GLCPs). Moreover, we introduce cooperative color and GLCP features for neighborhood matching in the synthesis process. We assign different weights for color and grey-level features according to the characteristics of the input texture. The results show that the proposed approach performs well in terms of the synthesis quality.

Pollen grains vary in colour and shape and can be detected in honey used as a way of identifying nectar sources. Accurate differentiation between pollen grains record is hampered by the combination of poor taxonomic resolution in pollen identification and the high species diversity of many families. Pollen identification determines the origin and the quality of the honey product, but this indefiniteness is also a big challenge for the beekeepers. This study aimed to develop effective, accurate, rapid and non-destructive analysis methods for pollen classification in honey. Ten different pollen grains of plant species were used for the estimation. GLCM (grey level co-occurrence matrix) texture features and ANN (artificial neural network) were used for the identification of pollen grains in honey by the reference of plant species pollen. GLCM has been calculated in four different angles and offsets for the pollen of the plant and the honey samples. Each angle and offset pair includes five features. ...

This paper discusses on how the Kohonen Self Organizing Map (KSOM) is used as a tool to cluster and classify the tropical wood species. Wood features have been extracted through the use of two features extractors; Basic Grey Level Aura Matrix (BGLAM) and Statistical Properties of Pores Distribution (SPPD) techniques from the wood images. The wood dataset is trained and tested separately using KSOM algorithm with different parameters such as the number of epochs and map sizes in order to find the best topological network for clustering and classifying the wood data. The clustering results are analyzed and the best result is selected based on common KSOM performance measurement; topological error and quantization error. The number of cluster performed by KSOM is 61 clusters, while the number of overlapped cluster varies for each map. From the results, the 23x23 map size has produced the lowest number of overlapped clusters with the minimum value of topological error and quantization error.

Tomato yellow leaf curl disease spreads very fast and often causes huge yield losses. It usually represents obvious symptoms on leaves at least 15 days after being infected. This paper investigated the possibility of discriminating tomato yellow leaf curl disease by hyperspectral imaging technique. A hyperspecral imaging system was used to collect hyperspectral images of healthy and diseased tomato leaves. The edge and midrib of leaves were chosen as regions of interest (ROI). Reflectance spectra in the range of 450-1000 nm of ROIs of healthy and infected leaves were measured. The raw, first-derivative, and absolute difference spectra were analyzed to select the feature wavelength for discriminating diseased and healthy leaves. Mean grey values and eight texture features of diseased and healthy leaves images at feature wavelength extracted by grey level co-occurrence matrix (GLCM) were analyzed by receiver operator characteristic curve (ROC). Nine features in total were analyzed. The results showed leaf edge was more prone to disease than leaf midrib area. Red edge shifting in the range of 695-750 nm and blue shifting in the range of 515-560 nm presented in the first-derivative reflectance. The highest difference value of diseased and healthy leaves was in the rang of 710-730 nm. The image at 853 nm was used for segmenting leaf area out and the image at 720 nm was used for acquiring grey information of leaves. ENT_DEV, ENT_MEAN, INE_DEV features were the top three for their areas under the curves (AUC) were more than 0.9.

International audience; Image segmentation techniques have been widely used in medical image analysis. However, existing methods can not provide exact physical significance of segmented image regions because they are mainly based on basic image features such as grey level and texture without taking into account specialised medical knowledge. However, medical knowledge plays an indispensable role when doctors analyse medical images in their diagnosis. To deal with this problem, many tissue classification systems were developed by incorporating specific medical knowledge. However, these systems strongly depend on specific applications and then can not provide a general structure for integrating medical knowledge in a larger context. To treat medical image recognition in a systematic way, we propose in this paper a general intelligent tissue classification system which combines both FCM (fuzzy C-means) based clustering algorithm and qualitative medical knowledge on geometric properties of different tissues. In this system, a general model is proposed for formalising non structured and non normalised medical knowledge from various medical images. This model utilises a DOGMA approach (a natural language-based ontology system) for formal representation of these geometric features.

It has long been accepted that the texture of a froth is generally a good qualitative indicator as to the performance of the flotation process. During investigation into quantifying froth textures by image analysis, it was observed that the grey level of a coal froth can provide significant information. The grey level showed correlations with flotation performance (e.g. ash content and froth mass flows rates) for both a single flotation cell and a bank of four pilot-scale cells. However, the two systems showed distinct differences in the texture of the froth surfaces. Testwork demonstrated that the effect of varying background illumination had a negligible effect on the results.

This paper investigates the suitability of using airborne SAR images from the GlobeSAR mission for monitoring of growth stages of paddy crops. A large paddy plantation area in Kedah state, Malaysia, was selected for the study. Simple ground‐truth measurements were conducted at the same time as the flight mission. In general, the paddy growth pattern can be divided into three major stages, and the preliminary study of the acquired single channel grey level SAR images indicates that the GlobeSAR images have a great potential for monitoring of the different growth stages of paddy crops. Future work will further explore microwave remote sensing techniques for paddy crop monitoring using images from RADARSAT and JERS‐1.

Summary An interactive semi-automatic procedure for extraction of cell contours from light microscope images is presented. The user is required to specify four contour points and the algorithm determines the rest of the contour automatically. The algorithm exploits the fact that cell contours have lower grey level than their immediate surrounding and are usually very similar in shape to some piece-wise ellipses. A cost function is defined to detect the cell contours incorporating both the elliptical shape and local image intensities. The procedure is fast, reliable and well suited for routine interactive applications.

Point distribution models (PDMs) allow a compact description of an object's shape to be found from a set of example images. In previous work by the second author a method of incorporating grey level information into a PDM was developed. Further work investigated fitting such a composite model to image data consisting of a set of images of a pig viewed from above. This paper describes work on images containing more than one pig. A technique for initialising the model is used which searches the image for ridges in the grey level landscape. These generally lie along the backbone of the animal and provide a good starting point for automatic fitting. By minimising an objective function which measures the difference in grey level and the error in boundary correspondence, an accurate fit of model to data is obtained. Ridge detection initialises the model to within ±12 pixels of the object. Strict limits on the boundaries of the search space constrain the minimisation process allowing convergence to the true minimum. The resulting fit is good even for objects which are partially obscured. Poor final values of the objective function allow the detection of erroneous results.

OBJECTIVE To assess the prognostic value of morphonuclear parameters determined by means of computerized image analysis in untreated bladder tumours. PATIENTS AND METHODS Fifty patients had untreated bladder cancer; in 28 patients the tumours were superficial (7pT1G1; 14pT1G2; 7pT1G3) and in 22 they were invasive (6T2G2, 9T2G3, 2T3G2, 5T3G3). Feulgen-stained imprints were processed for morphonuclear analysis on a SAMBA 200 computerized image analysis system (T1TN, France), which measures optical density (integrated optical density, IOD; surface area, SURF; mean optical density, MOD), texture (long run length, LRL; short run length, SRL; run length distribution, RLD; run length percentage, RLP; grey level distribution, GLD) and contrast (contrast, C; energy, E). RESULTS Morphonuclear parameters IOD, SURF, LRL, SRL and C were found to correlate with the risk of recurrence and progression of superficial bladder tumours (results for ANOVA respectively IOD P < 0.001; SURF P = 0.02; LRL P = 0.05; RLD P = 0.04; SRL P = 0.04; C P < 0.001). In invasive bladder cancer, parameters IOD and C only correlated with the risk of progression (ANOVA respectively IOD P < 0.001; C P < 0.001). On the other hand, progression-free curve analysis using the Kaplan Meier method showed that morphonuclear parameters may be useful in predicting the outcome for superficial tumours (Log/Rank test: SURF P < 0.001; RLD < 0.001; SRL = 0.003; LRL < 0.001; C < 0.001; IOD < 0.001). In the case of invasive tumours, only two parameters provided prognostic information (Logrank test: C < 0.001; IOD < 0.001). CONCLUSION Nuclear morphometry assessed by image analysis is potentially useful in assessing the prognosis of bladder tumours; it provides objective and quantitative parameters. Further studies will determine whether morphonuclear analysis can be used to monitor the treatment of bladder cancer, particularly superficial tumours.

Second harmonic generation microscopy (SHGM) was used to monitor the process of chronological aging skin in vivo. The collagen structures of mice model with different ages were obtained using SHGM. Then, texture feature with contrast, correlation and entropy were extracted and analysed using the grey level co-occurrence matrix. At last, the neural network tool of Matlab was applied to train the texture of collagen in different statues during the aging process. And the simulation of mice collagen texture was carried out. The results indicated that the classification accuracy reach 85%. Results demonstrated that the proposed approach effectively detected the target object in the collagen texture image during the chronological aging process and the analysis tool based on neural network applied the skin of classification and feature extraction method is feasible.

The average grey-levels (AVGL) calculated from B-Mode images were assessed for non-invasive temperature estimation in a porcine kidney sample. The porcine kidney was subjected to heating and cooling procedure with temperature varying from 36°C to 45°C. The sample was continuously imaged with an ultrasound scanner, and simultaneously temperature was measured at each 5 s using two Type-T thermocouples placed at the central region and at the bottom of the sample. The result shows that AVGL/Temperature relations were highly correlated for both regions, and considering heating and cooling periods. Besides, AVGL/Temperature functions estimated fitting errors inferior to ± 0.25°C, indicating that it might be possible to track temperature changes from both tissues using AVGL. This result is innovative since it suggests that AVGL seems to be robust to estimate non-invasively temperature variations in B-mode images from tissues presenting a complex morphological structure.

La técnica de Análisis Digital de Imagen (ADI) se basa en la utilización de un equipo de ADI compuesto por un microscopio de refl exión, una cámara de video 3CCD, una tarjeta digitalizadora y un software de análisis de imagen. Todos estos elementos se interrelacionan de manera que las escenas visualizadas en el microscopio son capturadas con la cámara de video y digitalizadas y almacenadas en el ordenador, donde, a través del software de análisis de imagen serán tratadas para obtener la información necesaria. Para que esta información sea fi able y reproducible es necesario que el equipo trabaje en condiciones estables. E. Berrezueta (2004) desarrolló una metodología de trabajo con el equipo de ADI del Laboratorio de Microscopia Aplicada de la Escuela Técnica Superior de Ingenieros de Minas de Madrid, que ha sido ensayada para constatar cómo un usuario sin experiencia previa es capaz de lograr la puesta a punto del equipo de ADI, obteniendo resultados fi ables y reproducibles. La validez del método ha sido comprobada mediante la aplicación de ADI a la cuantifi cación mineral de probetas pulidas estudiadas previamente por un mineralogista experto, y cuantifi cadas por el método tradicional del contador de puntos., The equipment of DIA is composed by a refl ection microscope, a video camera 3CCD, a frame grabber and the analysis image software. All these elements are interrelated so that the scenes visualized in the microscope are captured by the video camera and digitized and stored in the computer, where, through the analysis image software they will be processed to obtain the necessary information. The methodology to be tested is the one developed by E. Berrezueta (2004), in the Laboratory of Applied Microscopy of the Superior School of Mines, Madrid, where the present work has been carried out. The above mentioned methodology describes the necessary steps to achieve the adjustment of the equipment of DIA, that is to say, to reach the necessary conditions of stability to obtain reliable and reproducible results. One of the above mentioned conditions are reached, the validity of the method will be tested by means of its application to the mineral quantifi cation of polished samples studied before by an expert mineralogist, and quantifi ed by the traditional method of the points counter operator.

The equipment of DIA is composed by a refl ection microscope, a video camera 3CCD, a frame grabber and the analysis image software. All these elements are interrelated so that the scenes visualized in the microscope are captured by the video camera and digitized and stored in the computer, where, through the analysis image software they will be processed to obtain the necessary information. The methodology to be tested is the one developed by E. Berrezueta (2004), in the Laboratory of Applied Microscopy of the Superior School of Mines, Madrid, where the present work has been carried out. The above mentioned methodology describes the necessary steps to achieve the adjustment of the equipment of DIA, that is to say, to reach the necessary conditions of stability to obtain reliable and reproducible results. One of the above mentioned conditions are reached, the validity of the method will be tested by means of its application to the mineral quantifi cation of polished samples studied before by an expert mineralogist, and quantifi ed by the traditional method of the points counter operator. La técnica de Análisis Digital de Imagen (ADI) se basa en la utilización de un equipo de ADI compuesto por un microscopio de refl exión, una cámara de video 3CCD, una tarjeta digitalizadora y un software de análisis de imagen. Todos estos elementos se interrelacionan de manera que las escenas visualizadas en el microscopio son capturadas con la cámara de video y digitalizadas y almacenadas en el ordenador, donde, a través del software de análisis de imagen serán tratadas para obtener la información necesaria. Para que esta información sea fi able y reproducible es necesario que el equipo trabaje en condiciones estables. E. Berrezueta (2004) desarrolló una metodología de trabajo con el equipo de ADI del Laboratorio de Microscopia Aplicada de la Escuela Técnica Superior de Ingenieros de Minas de Madrid, que ha sido ensayada para constatar cómo un usuario sin experiencia previa es capaz de lograr la puesta a punto del equipo de ADI, obteniendo resultados fi ables y reproducibles. La validez del método ha sido comprobada mediante la aplicación de ADI a la cuantifi cación mineral de probetas pulidas estudiadas previamente por un mineralogista experto, y cuantifi cadas por el método tradicional del contador de puntos.

The average grey-levels calculated from B-Mode images were assessed for non-invasive temperature estimation in a porcine tissue sample containing two different tissue types, fat and muscle. The porcine sample was subjected to heating and cooling procedures with temperature varying from 35°C to 42°C. The sample was continuously imaged with an ultrasound scanner, and simultaneously the temperature was measured at each 5 seconds using Type-T thermocouples. The result shows that the average grey-level (AVGL)/temperature relations are different for the two studied regions, where the muscle tissue tends to present a bigger AVGL variation than the fat tissue considering the same temperature variation. Besides, the average grey-level/temperature functions estimated for each tissue region presented fitting errors inferior to ± 0.21°C, indicating that it might be possible to track temperature changes from both tissues using AVGL. This result is innovative since it suggests that using the same B-mode image and an average grey-level/temperature function to each region it is possible to estimate non-invasively temperature variations from different tissue regions in the same tissue sample. Future work includes the investigation of the spatial limits of these average grey-level/temperature functions.

The noninvasive diagnosis of the malignant tumors is an important issue in research nowadays. Our purpose is to elaborate computerized, texture-based methods for performing computer-aided characterization and automatic diagnosis of these tumors, using only the information from ultrasound images. In this paper, we considered some of the most frequent abdominal malignant tumors: the hepatocellular carcinoma and the colonic tumors. We compared these structures with the benign tumors and with other visually similar diseases. Besides the textural features that proved in our previous research to be useful in the characterization and recognition of the malignant tumors, we improved our method by using the grey level cooccurrence matrix and the edge orientation cooccurrence matrix of superior order. As resulted from our experiments, the new textural features increased the malignant tumor classification performance, also revealing visual and physical properties of these structures that emphasized the complex, chaotic structure of the corresponding tissue.

This paper presents a novel method for the classification of rocks into the three major categories, namely, igneous, sedimentary and metamorphic. Each of these rock types has various sub-types too. The various Tamura Features are formulated and calculated from the input image. The values obtained are compared with the query image by Sum of Squared Distance (SSD. The classified results are then compared with those results of Grey Level Cooccurance Matrix (GLCM), Color cooccurance matrix and Moments. The proposed method outperforms the other previously developed methods by providing the classification accuracy of more than 87% for all the three types of rocks. The proposed method significantly improves efficiency with less computational complexity.

National audience; Reducing sodium intake in Western diet is advised for reducing hypertension and risk of developing cardiovascular disease. Sodium is mainly consumed via processed foods and among them breads are one of the most important sources of salt [1]. The present study investigated the changes in mastication and salivation for bread samples with different composition and texture, their respective contribution to food bolus formation, and the impact on salt release. The study set-up included five subjects presenting different chewing efficiency and four different breads from different composition and structure (two French baguettes (bakery (BB) and supermarket (BS)), a toast bread (TB) and a German rye bread (RB)). Samples were chewed for 10, 20, 30 chewing cycles (C10, C20 and C30) and just before swallowing (SW, corresponding to the natural consumption habits). Chewing behavior was determined using surface electromyography and salivation was quantified from the water contents of the collected boluses. The kinetic of bread degradation during food bolus formation was characterized using images texture analysis. Images from collected boluses were analyzed using the Grey level co-occurrence matrix (GLCM) method. The textural feature contrast was selected as markers of food degradation [2]. The kinetic of salt release was determined at the same chewing periods. Saliva was collected by spitting and analyzed by high-performance liquid chromatography. Mastication and salivation varied both between products and between subjects. The chewing activity and number of chewing cycles necessary for preparing a swallowable bolus (SW) were the highest for the baguettes. Products also induced different bolus lubrication: water content of boluses was the highest for the bakery baguette and the lowest for the toast bread. These data showed an adaptation in chewing behavior as function of bread characteristics (humidity, proportion crust/crumb, density of the crumb and textural properties of the crust). Along the chewing sequence, bread boluses gradually gained in homogeneity (inducing a decrease in contrast). As a consequence of the different chewing behaviors previously described, the rate of food bolus formation varied both between breads and between subjects. After 20 to 30 chewing cycles TB and RB boluses presented homogeneity close to those of swallowable boluses, whereas the swallowing threshold for the baguettes was still not achieved after C30. The analysis of contrast also revealed specific patterns of food degradation between subjects, which could be explained by their mastication behaviour and chewing efficiency. The observed differences in bread oral processing significantly impacted the kinetic of salt release in the mouth. Salt release was predicted by both saliva produced and bolus homogeneity. All together theses experiments showed that mechanisms of mastication and salivation are adapted to the nature of the food. They highly vary among subjects and specific profiles of bread degradation were observed. These profiles clearly impact the quantity of salt release in the mouth.

PURPOSE Evaluation of image quality of paper prints in routine CT investigations. METHOD The image quality of paper prints from 104 CT scans of daily routine investigations was analysed by three observer three observers according to a five-point ranking scale for contrast, grey level, spatial resolution, and subjective confidence in correct evaluation. Each study was rated "acceptable" or "not acceptable for documentation". RESULTS In 312 ratings the mean grade for contrast was 2.7, for grey levels 3.2, and for spatial resolution 3.3. Grades ranged from 1 = very good to 5 = insufficient. Subjective confidence in correct evaluation was rated as "certain" in 77.4 %, "likely" in 18.4 % and "impossible to evaluate" in 4.2 % of cases. 93.7 % of printed CT images were rated "acceptable for documentation". CONCLUSION Image quality of paper prints does not reach the quality of laser films concerning its contrast, grey levels, and spatial resolution. Nevertheless, paper prints are acceptable for documentation of findings in most CT investigations. A high confidence in correct evaluation was found. Limitations are found in documentation of small coin lesions of the lung and ischemic lesions of the neurocranium.

A method for writer-independent off-line handwritten signature verification based on grey level feature extraction and Real Adaboost algorithm is proposed. Firstly, both global and local features are used simultaneously. Secondly, dissimilarity vector is adopted. Finally, Real Adaboost algorithm is applied. Experiments on the public signature database GPDS Corpus show that our proposed method has achieved the FRR 5.64% and the FAR 5.37% which are the best so far compared with other published results.

The background grey level is an important facto r affecting the precision of grey system models. To improve the precision of the grey model, the background value was rebuilt based on the monotonic rise of accumulated generating sequence. The sum of triangle area subtracted from narrow trapezium area is instead of the tradition background value. The rebuilt background value w as more close to the practical value, and the model accuracy was improved. Finally, the step of grey modeling was given, and the application result s showed that it is effective.

This paper presents a new contour extraction algorithm for time-of-flight (ToF) camera distance images. Experimental results and comparisons with the classical grey level contour extractors are presented. A mathematical model for contour validation is developed, and finally, an edge alignment method is proposed.

Pneumoconiosis is diagnosed as categories 0-4 according to the Pneumoconiosis Law. Physicians have difficulty precisely categorizing many chest images. Therefore, we have developed a computerized method for automatically categorizing pneumoconiosis from chest radiographs. First, we extracted the rib edge regions from lung ROIs. Second, texture features were extracted using a dot enhancement filter, line enhancement filter, and grey level co-occurrence matrix. Third, the rib edge regions were removed from these processed images. Finally, we used a support vector machine for feature analysis. In a consistency test, 56 cases (69.7%) were classified correctly, and 45 cases (61.8%) were classified correctly in a validation test. These results show that the proposed features and removal of the rib edge are effective in classifying the profusion of opacities that indicate pneumoconiosis.

Introduction: To minimize fatigue due to eye adaptation and maximize contrast perception, it has been suggested that lesion luminance be matched to overall image luminance to perceive the greatest number of grey level differences. This work examines whether lesion display luminance matches the overall image and breast tissue display luminance and whether these factors are positioned within the optimum luminance for maximal contrast sensitivity. Methods: A set of 42 mammograms, collected from 21 patients and containing 15 malignant and 6 benign lesions, was used to assess overall image luminance. Each image displayed on the monitor was divided into 16 equal regions. The luminance at the midpoint of each region was measured using a calibrated photometer and the overall image luminance was calculated. Average breast tissue display luminance was calculated from the subset of regions containing of only breast tissue. Lesion display luminance was compared with both overall image display luminance and average breast tissue display luminance. Results: Statistically significant differences (p

This paper proposes an automated system for recognizing disease conditions of human skin in context to health informatics. The disease conditions are recognized by analyzing skin texture images using a set of normalized symmetrical Grey Level Co-occurrence Matrices (GLCM). Directional GLCMs are computed along four directions viz. horizontal, vertical, right diagonal, left diagonal, and a set of features viz. Contrast, Homogeneity, Mean, Variance and Energy computed from each, are averaged to provide an estimation of the texture class. The system is tested on a set of medical images displaying three dermatological skin conditions viz. Acne, Eczema, and Urticaria. The features are considered in various combinations viz. individually, and in joint 2-D feature spaces, using L1 and L2 metrics as well as neural network classifiers, to study which combination produces best recognition accuracies.

This paper presents a novel method for the automatic pupil and iris localization. The proposed algorithm is based on an automatic adaptive thresholding method that iteratively looks for a region that has the highest chances of enclosing the pupil. Once the pupil is localized, next step is to find the boundary of iris based on the first derivative of each row of the area within the pupil. We have tested our proposed algorithm on two public databases namely: CASIA v1.0 and MMU v1.0 and experimental results show that the proposed method has satisfying performance and good robustness against the reflection in the pupil.

International audience; This chapter presents an application of a stigmergic approach to extract regions in grey-level images. This application is based on the model of a social spiders behaviour which has been presented earlier in Chap. 6. This chapter first introduces the region detection problem, justifies the interest of a multi-agent application for this issue, presents the transposition of the spider model and shows the results obtained by this approach.

A video-imaging technique, using commercial software to process images obtained at 550 nm, was established to estimate chlorophyll content of cucumber fruit disks. The chlorophyll content of excised disks was extracted, determined, and regressed on the video-image grey level. They were linearly related. The change in grey level of the whole visible image accurately indicated the change of green color during fruit development on the vine and the loss of green color after 1 week of storage at 13C. The relationship of the chlorophyll content on grey level was quadratic for three imaging methods: 1) average grey level of the five disks; 2) average grey level of the whole cucumber image; and 3) average grey level of central one-third of the whole cucumber image. Chlorophyll content was most highly correlated to the grey level of the disks themselves (residual SD = 6.74 μg·cm-2), but this sampling technique was destructive. Both one-third of the fruit image (SD = 9.25 μg·cm-2) and the whole image (SD = 9.36 μg·cm-2) provided satisfactory precision. For simplicity, whole-fruit imaging is suitable for estimating fruit chlorophyll content and for quantifying fruit green color intensity. Potential use of this technique in product sorting and shelf life prediction of long English cucumbers is discussed.

A class of scientific images, which we will call blot images, contains information in the form of relationships between grey level pixels. One way to extract this information is to fit model functions to the objects in the image. We have explored the use of a Moffat function as a data model, and use the technique of simulated annealing to fit many instances of this function to the data in the image. Two examples are presented: stellar photometry, a natural application for the Moffat function, and reading DNA sequencing gels.

An automatic procedure to estimate proportions of components from grey level histograms is proposed. This procedure is based on statistical methods for parameter estimation in mixtures of normal distributions by maximum likelihood. The major advantage is that proportions of components can be estimated properly even when the grey level distributions of the components overlap considerably.

In three-dimensional surface display, the grey-level values of the voxels on the surface of an object are not always constant. This fact means that the traditional grey-level gradients are not the fact representation of the surface normals. The author presents an improved method of estimation of surface normal for 3D surface display. The new method calculates gradient at the location where the grey-level is equal to threshold value. Using this scheme, the gradients are estimated on an isogrey-level curve and are the exact surface normals. The images generated using this method show very fine details. The validity of the proposed scheme is illustrated with a number of examples from medical imaging. The results are compared to images rendered by the traditional grey-level gradient method. The proposed algorithm has advantages over the traditional method in improved estimation of surface normal and rendering of finer details.

OBJECTIVE: To evaluate the effect of casein phosphopeptide–amorphous calcium phosphate tooth mousse on the remineralization of bovine incisor by circularly polarized images. METHODS: Eighty bovine incisors, each with a 4 × 4 mm artificially demineralized area, were used. The samples were divided into four groups: Group A, casein phosphopeptide–amorphous calcium phosphate tooth mousse; Group B, fluoride toothpaste; Group C, casein phosphopeptide–amorphous calcium phosphate tooth mousse and fluoride toothpaste; and Group D, no treatment. Circularly polarized images were taken after the specimens were treated for 3, 6, 9, or 12 weeks, and the size of the demineralized area and the mean grey level were measured. Data analysis was done using repeated measures variance analysis. Pearson correlation coefficients were computed to evaluate the correlation between the size of the demineralized area and the mean grey level. RESULTS: In all four groups, the size of the demineralized area and the mean grey level declined with time. The size of the demineralized area of Group C was significantly smaller than that of Group A at the end of the third and sixth weeks (P = .039, P = .000, respectively), and the mean grey level of Group C was lower than that of Group A at the end of the 6th and 12th weeks (P = .037, P = .004, respectively). At the end of the 6th, 9th, and 12th weeks, the size of the demineralized area of Group C was smaller (P = .000, P = .005, P = .005, respectively) and the mean grey level was lower (P = .000) than those of Group B. No statistically significant correlations were detected between the size of the demineralized area and the mean grey level. CONCLUSION: Casein phosphopeptide–amorphous calcium phosphate tooth mousse can reduce the size and mean grey level of demineralized areas and promote the remineralization of bovine enamel. Combined application with fluoride toothpaste strengthens the effect.

Image multi-threshold segmentation method based on two-dimensional Tsallis entropy is proposed by utilizing Tsallis entropy. The improved particle swarm optimization is used to search best two-dimensional multi-threshold vectors by maximising the two-dimensional Tsallis entropy. The proposed method not only considers the spatial information of pixels, but also the interaction between the object and background, the different responses in variant grey level. The experimental results show that the new algorithm is better than the tradition methods with both a better stability and a higher speed.

This paper proposes an automated system for recognizing disease conditions of human skin in context to medical diagnosis. The disease conditions are recognized by analyzing skin texture images using a set of normalized symmetrical Grey Level Co occurrence Matrices (GLCM). GLCM defines the probability of grey level i occurring in the neighborhood of another grey level j at a distance d in directionθ. Directional GLCMs are computed along four directions: horizontal (θ = 0°), vertical (θ = 90°), right diagonal (θ = 45°) and left diagonal (θ = 135°), and a set of features viz. Contrast, Homogeneity and Energy computed from each, are averaged to provide an estimation of the texture class. The system is tested using 225 images pertaining to three dermatological skin conditions viz. dermatitis, eczema, urticaria. An accuracy of 94.81% is obtained using a multilayer perceptron (MLP) as a classifier.

During the acquisition of a mammogram the breast is compressed between the compression paddle and the support table When compression is applied the upper plate is tilted which results in variation in breast thickness from the chest wall to the breast margin Variation in breast thickness influences the grey level values of the image and hampers image analysis, such as volumetric breast density estimation In this paper we present and compare two methods that estimate and correct image tilt The first method estimates tilt from fatty tissue regions The second method is based on the entropy of the grey level distribution of the image 1876 images are obtained from relatively young women with a high breast density on average The tilt correction methods are evaluated by assessing their accuracies in estimating artificial tilts that are added to the images that are expected to have only a small tilt On average both methods are able to estimate the artificial tilt, although the accuracy is relatively low To the best of our knowledge this is the first paper that presents and validate tilt correction methods on individual mammograms We expect that results will be better in screening populations which forms the majority of cases utilised in image analysis.

SUMMARY This paper presents an application of automatic image analysis to the morphometry of ceramic sheets. The sheets can be characterized by granulometrie and stereological texture parameters of the granular and the porous phases. Grey level granulometry, automatic thresholding and filtering have thus been used. The results show that the morphology of green sheets is strongly dependent on the ball-milling conditions.

An optical system for imaging the topography of (coating) surfaces using a video monitor is described. The grey level of the image correlates with the slope of the surface. The principle of operation of the method is explained and some applications are indicated.

A new experimental method called retrographic imaging has been developed which uses retroreflected light for visualising minute irregularities in reflective opaque surfaces. This paper provides a qualitative explanation of this phenomenon, the experimental setup required for such imaging, and describes an experiment which was devised to determine how this type of imaging can be correlated to actual surface topography. Results indicate that grey level intensity is proportional to an average between the surface contour and its rate of change of slope.

The current US peanut grading system is labour intensive and requires human judgement at several steps in the grading process. One goal of the peanut industry is to ensure high quality peanuts by increasing grading accuracy. Developing an automated grading system will assist in attaining that goal. The development of this grading system centers around a central system which automatically controls the grading process. The system makes decisions on single objects from a sample of peanuts. A machine vision system acquires an image of the object. Kernel size is determined from the image by the system. The kernel is evaluated for damage using grey level, tristimulus and texture information, and the appropriate sorting action is taken. Other grading devices which provide inputs into the automated grading system are a single kernel moisture meter, a bulk moisture meter, a bar code reader and a chemical testing system. Each of these devices are attached to the central processing unit and provide input into the automated grading system. After the entire sample has been processed, the categories are automatically weighed and the system determines the peanut quality based on the grade values. This automated system will result in a less labour-intensive system which will ensure the quality of US peanuts for domestic and foreign buyers and consumers worldwide.

In this article, we present a new method to extract internal and external borders (intimal-adventitial) of arteries from Optical Coherence Tomography (OCT) images. The method is based on A-scan segmentation. First, the distribution of the grey level values on every A-scan is analyzed separately using a sliding window to approximate a single-lobe distribution. Our hypothesis is that the position of the arterial tissue corresponds to the window which exhibits the largest single-lobe distribution. Once all the tissue is extracted from the image, every segmented A-scan position is corrected using a block of neighbouring segmented A-scans. Experimental results show that the proposed method is accurate and robust to extract arterial tissue., Series: Lecture Notes in Computer Science; no. 5627

In this work we review three methods proposed to estimate the fraction of alive sperm cells in boar semen samples. Images of semen samples are acquired, preprocessed and segmented in order to obtain images of single sperm heads. A model of intracellular density distribution characteristic of alive cells is computed by averaging a set of images of cells assumed to be alive by veterinarian experts. We quantify the deviation of the distribution of a cell from this model and use it for classification deploying three different approaches. One is based on a decision criterion used for single cell classification and gives misclassification error of 20.40%. The other two methods are focused on estimating the fraction of alive sperm in a sample, instead of single cell classification. One of them applies the least squares method, achieving an absolute error below 25% for 89% of the considered sample images. The other uses an iterative procedure to find an optimal decision criterion that equalizes the number of misclassifications of alive and dead cells. It provides an estimation of the fraction of alive cells that is within 8% of its actual value for 95% of the samples.