As today information era of advanced and secure digital technology field, monitoring system and security mechanisms are played as the most important role. By using specialized security camera in public sectors and pedestrian crossings, it can monitor and record a real time events and information of the sectors as video clips to track criminals. According to get the important data clearly and correctly from the video clips, the detection and extraction methods are essential. The proposed system focuses on the detection and extraction of car number plate that are taken from over speed driving cars. So, these number plates are deburred to overcome some of the security threat and enhance the motion deburring technique. Our proposed method is the combination of connected component based approach with the regional geometrical features. In this method, key frames are generated from an input video clips using Discrete Wavelet Transform (DWT) based approach. From the key frame images, rectangle shape areas which has high luminance value is detected and extracted as foreground regions and others are discarded as background by using regional geometric features. Finally, the rectangle shapes are checked whether any text is included or not. If a rectangle shape area contains text, this system accepts that it is a number plate and other region is omitted. Then the accuracy of the research method is evaluated with various experiments to compare with previous researches.
14 Read more
fundamental challenges, it poses, ﬁnds application in areas, such as human-computer interaction, clinical psychology, lie detection, pain assessment, and neurology. Generally the approaches to FER consist of three main steps: face detection, feature extraction and expression recognition. The recognition accuracy of FER hinges immensely on the relevance of the selected features in representing the target expressions. In this article, we present a person and gender independent 3D facial expression recognition method, using maximum relevance minimum redundancy geometrical features. The aim is to detect a compact set of features that suﬃciently represents the most discriminative features between the target classes. Multi-class one-against-one SVM classiﬁer was employed to recognize the seven facial expressions; neutral, happy, sad, angry, fear, disgust, and surprise. The average recognition accuracy of 92.2% was recorded. Furthermore, inter database homogeneity was investigated between two independent databases the BU-3DFE and UPM-3DFE the results showed a strong homogeneity between the two databases.
participants, which is in the direction expected from the P-bias. The net effect (experimental – control) was -6.46 deg (mean) and -5.1 deg (median), and the probability of the distribution of observed scores on the null hypothesis that is 0.0106 (paired t-test, t(8)=3.32). Values of of ~ 5 deg are similar to those reported by Morgan et al., (2013) and were not significantly different between experimental and control conditions. Values of and are the same order of magnitude, as again is typical for geometrical biases such as the Muller-Lyer (Morgan , Hole, and Glennerster, 1990) with in this instance being greater. The bias was not significantly different from that of Experiment 1.1, in agreement with the findings of Weintraub, Kranz & Olson (1980), so we cannot exclude the possibility that a single pointer is sufficient for the full P-bias. This suggests that one origin of the P-effect is a mispointing or misangulation of the pointer, in agreement with findings and analysis of Ninio & O’Regan (1999) and Ninio (2014).
41 Read more
Nanoporous materials are of broad interest for various applications, in particular advanced thermoelectric materials. The introduction of nanoscale porosity, even at modest levels, has been known to drastically reduce a material’s thermal conductivity, in some cases even below its amorphous limit, thereby significantly increasing its thermoelectric figure of merit, ZT. The details of the important attributes that drive these large reductions, however, are not yet clear. In this work, we employ large-scale equilibrium molecular dynamics to perform an exhaustive atomistic-scale investigation of the effect of porosity on thermal transport in nanoporous bulk silicon. Thermal transport is computed for over 50 different geometries, spanning a large number of geometrical degrees of freedom, such as cylindrical pores and voids, different porosities, diameters, neck sizes, pore/void numbers, and surface-to-volume ratios, placed in ordered fashion, or fully disordered. We thus quantify and compare the most important parameters that determine the thermal conductivity reductions in nanoporous materials. Ultimately, we find that, even at the nanoscale, the effect of merely reducing the line-of-sight of phonons, i.e. the clear pathways that phonons can utilize during transport, plays the most crucial role in reducing the thermal conductivity in nanoporous materials, beyond other metrics such as porosity and surface/boundary scattering.
34 Read more
Data obtained measuring the release of fibrinogen from PDLLA films are shown in figure 4.6. A general consideration is that the release is not constant over all the test period. Graph shows a multi-step release, composed of periods where the protein is released faster and periods where the release rate is lower. Probably, this behavior is due to the complex geometry of these systems, in which pores structure plays a major role. Dimension, interconnectivity and structures of pores determine sample interaction with the liquid in which it is immersed, thus influencing protein desorption. The possibility of modulating this multi-step behavior can be used when a non-constant release of drugs is needed. Modification of the pores features leads to different behavior, leading to the possibility of tuning the release in multiple ways.
162 Read more
In this paper, we present a new supervised classification approach for target recognition in SAS images. The approach uses geometrical features and aims to make use of the increased image fidelity available in both target highlight and shadow response. The recognition procedure starts with a novel detection/segmentation stage based on the Hilbert transform , which partitions the image into highlights and shadow areas in order to estimate the most likely position of the target. A number of geometrical features are then extracted around the estimated target position, and are then used to classify the object against a previously compiled database of target and nontarget features.
This work focuses on perforated domains., i.e., domains with holes (or inclusions) that are smaller than the characteristic mesh width. In our case, this amounts to stating that diam(B i ) << diam(Ω), and in turn, this will imply that, in most cases, |˜ e| < h, which is pre- cisely the case not allowed in . Over the years many authors have proposed solutions to this problem. One alternative is the Composite FEM method, see, e.g., , where the geometrical features are included in the finite element space, thus proposing a method whose dimension does not necessarly depend on the number of geometrical inclusions, see also  for the application of the same idea to the Stokes problem, and  for an adaptive strategy associated to a discon- tinuous Galerkin version of this method. Alternatively, the geometrical features of the domain can be taken into account at the mesh generation step. This idea is at the basis of some recent developments on discontinuous Galerkin methods on general polyheadral meshes (see , and  for a recent review). Finally, it is interesting to mention the approach described in  (see also the references therein for an extensive review of this type of approach), where a multiscale problem on a domain with inclusions has been approximated using a multiscale finite element approach based on the enrichment of the Crouzeix-Raviart method with bubble functions.
17 Read more
Geometrical feature matching techniques are based on the computation of a set of geometrical features from the picture of a face. The overall configuration can be described by a vector which representing the position and size of the main facial features like eyes and eyebrows, nose, mouth, and an outline of face. The primary works on automated face recognition by using geometrical features was done in 1973. Their system achieved 75% recognition rate on a database of 20 people using two images per person, one as the model and the other as the test image. In 1993 R. Bruneli and T. Poggio, automatically extracted a set of geometrical features from the picture of a face, such as nose width and length, mouth position and chin shape. There were 35 features extracted form a 35 dimensional vector. The recognition was then performed with a Bayes classifier. They achieved recognition rate 90% on a database of 47 people. I.J. Cox el at. introduced a mixture-distance technique which achieved 95% recognition rate on a query database of 685 individuals. Each face was represented by 30 manually extracted distances. Reference  used Gabor wavelet decomposition to detect feature points for each face image which reduced the storage requirement for the database. Typically, 35-45 feature points per face were generated. Two cost values, the topological cost, and similarity cost, were evaluated. The recognition accuracy of the right person was 86% and 94% of the correct person's faces were in the top three candidate matches. In summary, geometrical feature matching based on precisely measured distances between features may be useful for finding matches in a large database. However, it will be dependent on the accuracy of the feature location algorithms. Disadvantage of current automated face feature location algorithms do not provide a high degree of accuracy and require considerable computational time.
The demand for quality of food products we consume is increasing day by day. As the literacy rate is increasing in India so is the need for quality of food products is increasing. India is the second largest producer of rice grains first being China. As the production of rice is increasing so is the demand for its quality. This demand for quality of food grains is increasing because some of the traders cheat the shopkeepers by selling poor quality food grains which contains the particles like stones, sand, leaf, broken and damaged seeds etc. This kind of low quality of rice is sold without being noticed even and moreover there is no special scheme to find such poor quality grains. Therefore it is been a problem for both consumers and sellers. Now a days we are using the chemical methods for the identification of rice grain seed varieties and quality.The chemical method used also destructs the sample used and is also very time consuming method. These can be avoided by using a machine vision or the digital image processing system. These method is a non destructive, very fast and cheap compared to the chemical method and also an attempt to overcomes the drawbacks of manual process.An image processing technique is applied to extract various features of rice grains and classifies the grains based on geometrical features. The images have been properly enhanced to reduce noise and blurring in image. Finally image has segmented by applying proper segmentation methods so that edges may be detected effectively and thus rectification of the image has been done. The collected features are then used in Neural Network system for classifying the rice granules.
nature while the right hand side is an algebraic expression. Such characteristics of geometrical features that have been produced are sequence of vertices, sequence of edges, sequence of faces and sequence of cubes. These characteristics appeared in Pascal Triangle and were used to prove the conjecture.
ABSTRACT: This paper proposes two algorithms (k-mean clustering and back propagation) for palm recognition of an individual using geometrical and texture features. Palmprint recognition being one of the important aspects of biometric technology is one of the most reliable and successful identification methods. Thus palmprint recognition is a very interesting research area. In this study algorithm is proposed to extract geometrical features and principal lines as texture feature and result is compared in both cases. The database is taken from www.coep.org.in and http://www4.comp.polyu.edu.hk/~csajaykr/IITD/Database_Palm.htm for processing. MATLAB with version 18.104.22.1689 (R2010a) 32 bits will be used to implement the proposed work. In the case of k-mean clustering result is 95.65% and recognition time is 0.10. On the other hand in BP-ANN result is 93.47% and recognition time is 0.8.
The starting point of the license plate location is to judge the license plate through the features of the car license plate area. Available license plate features include five aspects: (1) that the geometrical features of the license plate, that is the height, width and their proportions, are within the confines; (2) the form feature is that the license plate is in a rectangular frame and characters are arranged according to certain rules in the rectangular frame with intervals; (3) the gray distribution of car license plate area feature is that the horizontal lines through the license plate have a gray distribution of continuous peaks and troughs; (4) the horizontal or vertical projection characteristics of car license plate area present a continuous peaks and troughs distribution and (5), the spectrum processes the image by row or column DTF transformation and its diagram contains the location information of the license plate. According to the rules set in 2007 by the People’s Republic norm GA36 2007 mobile license plate standard in China, the basic characters of a vehicle license plate are as follows:
Classification of facial expression depends on deformation of face corresponds to each expression. The facial expression classified into seven types: happy, surprise, sad, anger, fear, disgust, and neutral. A lot of classifier is used for facial expression recognition such as Naïve Bayesian classifier, Neural Networks, Bayesian classifier, K-Nearest Neighbour, Hidden Markov Model (HMM), Linear Discriminant Analysis, Multi-class SVM, K-Means Clustering and its performance varies according to the recognition rate accuracy obtained for each facial expressions. NB classifier is used for facial expression recognition in . Wenming et al. introduces the Gabor wavelet transformation method which converts the 34 geometric points into a labelled graph vector. For facial expression recognition uses Kernel Canonical Correlation Analysis (KCCA) tested on Japanese female facial expression (JAFFE) .The recognition rate for six expression using KCCA achieved as 77.05%. Petar and Aggelos  extracts the facial animation parameter using active contour algorithm and Multistream Hidden Markov Model (MHMM) based automatic facial expression recognition have used. This technique achieves high recognition rate for fear and sadness facial expressions apart from another four expressions. Kotsia and Pitas  have proposed a method which is based on mapping and tracking of facial features from the video frame. They have used Candide wire frame model for feature tracking. Recognition is semiautomatic, in the sense means that the user has to manually place the candied grid nodes on facial landmarks depicted at the first frame of image sequence. Multi-class Support Vector Machine classifier classifies the six expression into its corresponding classes, neutral state is not considered here. It obtains high recognition rate accuracy. Hamid et al. proposed a fixed geometric model for geometric normalization of facial images. Multiclass Support Vector Machine with polynomial kernel representation for classifying the selected Extended Cohn-Kanade datasets. This eliminates geometric variability in emotion expressions and obtain high recognition rate for fear, surprise and happiness expressions with less computational cost . Viola-Jones Algorithm detects the region of interest face and extracts the features using image normalization and thresholding technique. Active Appearance Model (AAA)  tracks the facial feature points and classifies the seven facial expressions using Multiclass SVM. This method is experimented on Cohn- Kanade and IMM database.
The nonlinear geometrical parameter-pressure rela- tions of VSMC and nuclei, consistent with that of the outer diameter of vessels, suggest that recruited collagen fibers either in media or adventitia prevent the intact vessel as well as VSMC from overstretch at high pres- sure [12,15,20,21]. Additionally, the deformation of in- dividual VSMC was found to be affine. VSMC connect with the extracellular matrix (ECM) via focal adhesion and the deformation thus strongly depends on ECM de- formation as well as the macroscopic deformation of blood vessels. It is likely that flexible actin filaments deform with ECM through dense bodies in passive tissue such that collagen and elastin fibers follow affine defor- mation [22,23]. VSMC become stiffer during vasocon- striction due to the forces generated by actin-myosin in- teraction and tensile properties of cytoskeletal filaments increase significantly in contraction. Hence, the affine deformation assumption needs to be directly tested in active VSMC. Moreover, the elongation of the nuclei suggested that the tension developed by the cytoskeleton is transferred to the nuclei which may influence gene transcription and cellular phenotypes [14,24]. Conse- quently, the determination of strain and stress on indi- vidual VSMC is essential for better understanding of VSMC functions in normal and diseased arteries. This requires the development of microstructure based models to accurately predict the micro-environment of cells and nuclei.
The feature extraction algorithms based on the local patch, which could effectively obtain the local facial expression information, reduced the loss of facial expression information to a certain degree with respect to the geometric features. According to the 64 labeled key points, Wang et al.divided the face into 7 local patches, and used the distribution histogram of each surface to represent the human face expression changes, and achieved a recognition rate of 83.6%.
field of a purely advected quantity. The right-hand column shows the M function at the surface z = 31.3 km obtained from particle trajectories in full 3-D calculations. Similarly to the second column, the displayed information is Lagrangian, but here we obtain more fundamental information in this re- gard. This figure provides feedback for characterizing the time evolution of any purely advected scalar field, while the previous one displays just the realization of one particular initial datum. More specifically, the third column highlights the position and evolution of two hyperbolic points in the outer part of the vortex, as well as the vortex itself. As dis- cussed by García-Garrido et al. (2017), hyperbolic points are responsible for filamentation processes. Whether or not these filaments are eventually observed depends on the distribution of the scalar field. For instance, if the scalar field is com- pletely uniform in the whole domain, then its time evolution will show nothing about the features highlighted by M. How the features of the M field are visible in a scalar field depends on how the initial distribution of the advected field is with re- spect to the features of M. Figure 5 illustrates these facts in
14 Read more
The retina is a thin layer of cells at the back of the eyeball of vertebrates. It is the part of the eye which converts light into nervous signals. It is lined with special photoreceptors which translate light into signals to the brain. The main features of a fundus retinal image were defined as the optic disc, fovea, and blood vessels. Every eye has its own totally unique pattern of blood vessels. The unique structure of the blood vessels in the retina has been used for biometric identification and ophthalmology diagnosis.
Digital Images of lymphocytes were collected from various hospitals. 100 images are considered for the experimental purpose. 70 and 30 images are used for training and testing respectively for SVM classifier. 70 trainig images are supervised or labelled by the expert doctors as CB and non- CB. Fig 6(a) represents the output automated classification GUI in matlab for the input testing image with the results classified as CB and non-CB. In the Fig 6(b) the original image contains both centroblasts and non-centroblast cells. Fig 6 (c) show the classified output of the cells named as 01 as non-centroblast cell and 02 as centroblast cell. The count of CB and non-CB helps the doctors to decide the intensity of the disease. 84.53% accuracy has been achieved using SVM classifier with Log gabor and geometric features.
thickness, and initial mitral MR grade. Moreover, that study showed that patients with moderate MR were more likely to progress to severe MR, and that MR pro- gression results in an excess long-term complication rate, independent of confounding variables. Pini et al.  have shown that patients with MVP complicated by significant MR are more likely to have billowing and leaflet elongation than are MVP patients without MR. Although all features of mitral valve geometry alteration were not measured and excessive leaflet thickness was not specifically addressed in that study, those findings and those of our study are concordant. Accordingly, other studies have distinguished mitral valve billowing, in which leaflet apposition is normal, from MVP in which the leaflets fail to appose properly so that MR occurs [4,6,16]. Grayburn et al .  demonstrated that abnormal mitral leaflet coaptation on 2 dimensional echocardiography was strongly associated with the pre- sence of MR, with a prevalence of 71% (15 of 21 patients) and 20% (5 of 25 patients, p < 0.05) in patients with and without significant MR, respectively.
Abstract: In Computer vision system, rapidly expanding various applications. The goal in this paper is to develop a designing age classification system from the characteristics and information that can extract from the human face images for both sexes. The system proposed new algorithm that merging two features techniques (local and global) features. The local features including (primary face features), so the global features including (secondary face features). The new method in this paper present (local binary pattern) as a new technique uses in wrinkle analysis , so as this method uses to classify the input face images into one of four age groups: Baby, young, young adult and senior, and eight age categories: [1-6, 7-11, 12-19, 20-29, 30-39, 40-49, 50-65,66++]. This method based on human face region which contains a lot of information and properties that describe the head growth and face aging pretenses. These information can be used by the human brain to estimate the face age dependent on the external features that shows the craniofacial changes in geometrical characterize results by the growth of the head that changes the primary face features locations, the primary face features are: the center of the two eyes, nose peak, mouth peak, top head, face sides and the chin point, from these primary features we compute the geometrical ratios that distinguish babies faces from the three age groups: young, young adult and senior. The other changes that appear when the face aging is the texture changes which are the secondary features can be used to estimate the age of the face. The secondary face features may be the wrinkle appearance, duple chin, and eye bags. The wrinkle lines are calculated in the curliest five regions these are: for head, under two eyes and cheeks regions. These lines are computed and used to distinguish young, young adult, and senior age groups and age categories.
13 Read more