Abstract: Edge detection is usually the first step in medical
image processing. However, the difficulty increases when a
conventional kernel-based edge detector is applied to ultrasonic
images with a textural pattern and speckle noise. We designed an
adaptive diffusion filter to remove speckle noise while preserving the
initial edges detected by using a Sobel edge detector. We also propose
a genetic algorithm for edge selection to form complete boundaries of
the detected entities. We designed two fitness functions to evaluate
whether a criterion with a complex edge configuration can render a
better result than a simple criterion such as the strength of gradient.
The edges obtained by using a complex fitness function are thicker and
more fragmented than those obtained by using a simple fitness
function, suggesting that a complex edge selecting scheme is not
necessary for good edge detection in medical ultrasonic images;
instead, a proper noise-smoothing filter is the key.
Abstract: Face recognition in the infrared spectrum has attracted a lot of interest in recent years. Many of the techniques used in infrared are based on their visible counterpart, especially linear techniques like PCA and LDA. In this work, we introduce a probabilistic Bayesian framework for face recognition in the infrared spectrum. In the infrared spectrum, variations can occur between face images of the same individual due to pose, metabolic, time changes, etc. Bayesian approaches permit to reduce intrapersonal variation, thus making them very interesting for infrared face recognition. This framework is compared with classical linear techniques. Non linear techniques we developed recently for infrared face recognition are also presented and compared to the Bayesian face recognition framework. A new approach for infrared face extraction based on SVM is introduced. Experimental results show that the Bayesian technique is promising and lead to interesting results in the infrared spectrum when a sufficient number of face images is used in an intrapersonal learning process.
Abstract: This paper investigates the problem of automated defect
detection for textile fabrics and proposes a new optimal filter design
method to solve this problem. Gabor Wavelet Network (GWN) is
chosen as the major technique to extract the texture features from
textile fabrics. Based on the features extracted, an optimal Gabor filter
can be designed. In view of this optimal filter, a new semi-supervised
defect detection scheme is proposed, which consists of one real-valued
Gabor filter and one smoothing filter. The performance of the scheme
is evaluated by using an offline test database with 78 homogeneous
textile images. The test results exhibit accurate defect detection with
low false alarm, thus showing the effectiveness and robustness of the
proposed scheme. To evaluate the detection scheme comprehensively,
a prototyped detection system is developed to conduct a real time test.
The experiment results obtained confirm the efficiency and
effectiveness of the proposed detection scheme.
Abstract: There have been numerous implementations of
security system using biometric, especially for identification and
verification cases. An example of pattern used in biometric is the iris
pattern in human eye. The iris pattern is considered unique for each
person. The use of iris pattern poses problems in encoding the human
iris.
In this research, an efficient iris recognition method is proposed.
In the proposed method the iris segmentation is based on the
observation that the pupil has lower intensity than the iris, and the
iris has lower intensity than the sclera. By detecting the boundary
between the pupil and the iris and the boundary between the iris and
the sclera, the iris area can be separated from pupil and sclera. A step
is taken to reduce the effect of eyelashes and specular reflection of
pupil. Then the four levels Coiflet wavelet transform is applied to the
extracted iris image. The modified Hamming distance is employed to
measure the similarity between two irises.
This research yields the identification success rate of 84.25% for
the CASIA version 1.0 database. The method gives an accuracy of
77.78% for the left eyes of MMU 1 database and 86.67% for the
right eyes. The time required for the encoding process, from the
segmentation until the iris code is generated, is 0.7096 seconds.
These results show that the accuracy and speed of the method is
better than many other methods.
Abstract: The automatic construction of large, high-resolution
image vistas (mosaics) is an active area of research in the fields of
photogrammetry [1,2], computer vision [1,4], medical image
processing [4], computer graphics [3] and biometrics [8]. Image
stitching is one of the possible options to get image mosaics. Vista
Creation in image processing is used to construct an image with a
large field of view than that could be obtained with a single
photograph. It refers to transforming and stitching multiple images
into a new aggregate image without any visible seam or distortion in
the overlapping areas. Vista creation process aligns two partial
images over each other and blends them together. Image mosaics
allow one to compensate for differences in viewing geometry. Thus
they can be used to simplify tasks by simulating the condition in
which the scene is viewed from a fixed position with single camera.
While obtaining partial images the geometric anomalies like rotation,
scaling are bound to happen. To nullify effect of rotation of partial
images on process of vista creation, we are proposing rotation
invariant vista creation algorithm in this paper. Rotation of partial
image parts in the proposed method of vista creation may introduce
some missing region in the vista. To correct this error, that is to fill
the missing region further we have used image inpainting method on
the created vista. This missing view regeneration method also
overcomes the problem of missing view [31] in vista due to cropping,
irregular boundaries of partial image parts and errors in digitization
[35]. The method of missing view regeneration generates the missing
view of vista using the information present in vista itself.
Abstract: The paper describes a self supervised parallel self organizing neural network (PSONN) architecture for true color image segmentation. The proposed architecture is a parallel extension of the standard single self organizing neural network architecture (SONN) and comprises an input (source) layer of image information, three single self organizing neural network architectures for segmentation of the different primary color components in a color image scene and one final output (sink) layer for fusion of the segmented color component images. Responses to the different shades of color components are induced in each of the three single network architectures (meant for component level processing) by applying a multilevel version of the characteristic activation function, which maps the input color information into different shades of color components, thereby yielding a processed component color image segmented on the basis of the different shades of component colors. The number of target classes in the segmented image corresponds to the number of levels in the multilevel activation function. Since the multilevel version of the activation function exhibits several subnormal responses to the input color image scene information, the system errors of the three component network architectures are computed from some subnormal linear index of fuzziness of the component color image scenes at the individual level. Several multilevel activation functions are employed for segmentation of the input color image scene using the proposed network architecture. Results of the application of the multilevel activation functions to the PSONN architecture are reported on three real life true color images. The results are substantiated empirically with the correlation coefficients between the segmented images and the original images.
Abstract: For evaluating the severity of Chronic Obstructive Pulmonary Disease (COPD), one is interested in inspecting the airway wall thickening due to inflammation. Although airway segmentations have being well developed to reconstruct in high order, airway wall segmentation remains a challenge task. While tackling such problem as a multi-surface segmentation, the interrelation within surfaces needs to be considered. We propose a new method for three-dimensional airway wall segmentation using spring structural active contour model. The method incorporates the gravitational field of the image and repelling force field of the inner lumen as the soft constraint and the geometric spring structure of active contour as the hard constraint to approximate a three-dimensional coupled surface readily for thickness measurements. The results show the preservation of topology constraints of coupled surfaces. In conclusion, our springy, soft-tissue-like structure ensures the globally optimal solution and waives the shortness following by the inevitable improper inner surface constraint.
Abstract: The aim of this paper is to understand how peers can
influence adolescent girls- dieting behaviour and their body image.
Departing from imitation and social learning theories, we study
whether adolescent girls tend to model their peer group dieting
behaviours, thus influencing their body image construction. Our
study was conducted through an enquiry applied to a cluster sample
of 466 adolescent high school girls in Lisbon city public schools. Our
main findings point to an association between girls- and peers-
dieting behaviours, thus reinforcing the modelling hypothesis.
Abstract: Electric impedance imaging is a method of
reconstructing spatial distribution of electrical conductivity inside a
subject. In this paper, a new method of electrical impedance imaging
using eddy current is proposed. The eddy current distribution in the
body depends on the conductivity distribution and the magnetic field
pattern. By changing the position of magnetic core, a set of voltage
differences is measured with a pair of electrodes. This set of voltage
differences is used in image reconstruction of conductivity
distribution. The least square error minimization method is used as a
reconstruction algorithm. The back projection algorithm is used to
get two dimensional images. Based on this principle, a measurement
system is developed and some model experiments were performed
with a saline filled phantom. The shape of each model in the
reconstructed image is similar to the corresponding model,
respectively. From the results of these experiments, it is confirmed
that the proposed method is applicable in the realization of electrical
imaging.
Abstract: In the framework of adaptive parametric modelling of images, we propose in this paper a new technique based on the Chandrasekhar fast adaptive filter for texture characterization. An Auto-Regressive (AR) linear model of texture is obtained by scanning the image row by row and modelling this data with an adaptive Chandrasekhar linear filter. The characterization efficiency of the obtained model is compared with the model adapted with the Least Mean Square (LMS) 2-D adaptive algorithm and with the cooccurrence method features. The comparison criteria is based on the computation of a characterization degree using the ratio of "betweenclass" variances with respect to "within-class" variances of the estimated coefficients. Extensive experiments show that the coefficients estimated by the use of Chandrasekhar adaptive filter give better results in texture discrimination than those estimated by other algorithms, even in a noisy context.
Abstract: Speeded-Up Robust Feature (SURF) is commonly used for feature matching in stereovision because of their robustness towards scale changes and rotational changes. However, SURF feature cannot cope with large viewpoint changes or skew distortion. This paper introduces a method which can help to improve the wide baseline-s matching performance in term of accuracy by rectifying the image using two vanishing points. Simplified orientation correction was used to remove the false matching..
Abstract: The genre of fantasy depicts a world of imagine that triggers popular interest from a created view of world, and a fantasy is defined as a story that illustrates a world of imagine where scientific or horror elements are stand in its center. This study is not focused on the narrative of the fantasy, i.e. not on the adventurous story, but is concentrated on the image of the fantasy to work on its relationship with intended themes and differences among cultures due to meanings of materials. As for films, we have selected some films in the 2000's that are internationally recognized as expressing unique images of fantasy containing the theme of love in them. The selected films are 5 pieces including two European films, Amelie from Montmartre (2001) and The Science of Sleep (2005) and three Asian films, Citizen Dog from Thailand (2004), Memories of Matsuko from Japan (2006), and I'm a Cyborg, but That's OK from Korea (2006). These films share some common characteristics to the effect that they give tiny lessons and feelings for life with expressions of fantasy images as if they were fairy tales for adults and that they lead the audience to reflect on their days and revive forgotten dreams of childhood. We analyze the images of fantasy in each of the films on the basis of the elements of Mise-en-Scène (setting and props, costume, hair and make-up, facial expressions and body language, lighting and color, positioning of characters, and objects within a frame).
Abstract: Content-Based Image Retrieval (CBIR) has been
one on the most vivid research areas in the field of computer vision
over the last 10 years. Many programs and tools have been
developed to formulate and execute queries based on the visual or
audio content and to help browsing large multimedia repositories.
Still, no general breakthrough has been achieved with respect to
large varied databases with documents of difering sorts and with
varying characteristics. Answers to many questions with respect to
speed, semantic descriptors or objective image interpretations are
still unanswered. In the medical field, images, and especially
digital images, are produced in ever increasing quantities and used
for diagnostics and therapy. In several articles, content based
access to medical images for supporting clinical decision making
has been proposed that would ease the management of clinical data
and scenarios for the integration of content-based access methods
into Picture Archiving and Communication Systems (PACS) have
been created. This paper gives an overview of soft computing
techniques. New research directions are being defined that can
prove to be useful. Still, there are very few systems that seem to be
used in clinical practice. It needs to be stated as well that the goal
is not, in general, to replace text based retrieval methods as they
exist at the moment.
Abstract: In present paper we proposed a simple and effective method to compress an image. Here we found success in size reduction of an image without much compromising with it-s quality. Here we used Haar Wavelet Transform to transform our original image and after quantization and thresholding of DWT coefficients Run length coding and Huffman coding schemes have been used to encode the image. DWT is base for quite populate JPEG 2000 technique.
Abstract: This paper presents a novel method for prediction of
the mechanical behavior of proximal femur using the general
framework of the quantitative computed tomography (QCT)-based
finite element Analysis (FEA). A systematic imaging and modeling
procedure was developed for reliable correspondence between the
QCT-based FEA and the in-vitro mechanical testing. A speciallydesigned
holding frame was used to define and maintain a unique
geometrical reference system during the analysis and testing. The
QCT images were directly converted into voxel-based 3D finite
element models for linear and nonlinear analyses. The equivalent
plastic strain and the strain energy density measures were used to
identify the critical elements and predict the failure patterns. The
samples were destructively tested using a specially-designed gripping
fixture (with five degrees of freedom) mounted within a universal
mechanical testing machine. Very good agreements were found
between the experimental and the predicted failure patterns and the
associated load levels.
Abstract: In this article, we propose a methodology for the
characterization of the suspended matter along Algiers-s bay. An
approach by multi layers perceptron (MLP) with training by back
propagation of the gradient optimized by the algorithm of Levenberg
Marquardt (LM) is used. The accent was put on the choice of the
components of the base of training where a comparative study made
for four methods: Random and three alternatives of classification by
K-Means. The samples are taken from suspended matter image,
obtained by analytical model based on polynomial regression by
taking account of in situ measurements. The mask which selects the
zone of interest (water in our case) was carried out by using a multi
spectral classification by ISODATA algorithm. To improve the
result of classification, a cleaning of this mask was carried out using
the tools of mathematical morphology. The results of this study
presented in the forms of curves, tables and of images show the
founded good of our methodology.
Abstract: Based on the field investigation and long term remote
sensing data, the dynamics of the alpine wetland in the river basin and
their response to climate change were studied. Results showed the
alpine wetlands accounted for 3.73% of total basin in 2010. Lake and
river appeared an increasing trend in the past 30 years, with an
increase of 34.36 % and 24.57%. However, swamp exhibited a
tendency of decreasing with 233.74 km2. Annual average temperature,
maximum temperature, minimum temperature and precipitation in the
river basin all exhibited an increasing trend, whereas relative humidity
exhibited a decreasing trend. Ice and snow melting are main reasons of
lake and river area enhancement and swamp area descend. There
existed 91.78%-97.86% of reduced swamp converted into lakes on the
basis of remote sensing image interpretation. China-s government
policy of implementing development in the river basin is the major
driving force of artificial wetland growth.
Abstract: In this paper, an approach to reduce the computation steps required by fast neural networksfor the searching process is presented. The principle ofdivide and conquer strategy is applied through imagedecomposition. Each image is divided into small in sizesub-images and then each one is tested separately usinga fast neural network. The operation of fast neuralnetworks based on applying cross correlation in thefrequency domain between the input image and theweights of the hidden neurons. Compared toconventional and fast neural networks, experimentalresults show that a speed up ratio is achieved whenapplying this technique to locate human facesautomatically in cluttered scenes. Furthermore, fasterface detection is obtained by using parallel processingtechniques to test the resulting sub-images at the sametime using the same number of fast neural networks. Incontrast to using only fast neural networks, the speed upratio is increased with the size of the input image whenusing fast neural networks and image decomposition.
Abstract: This paper proposes a dual tree complex wavelet transform (DT-CWT) based directional interpolation scheme for noisy images. The problems of denoising and interpolation are modelled as to estimate the noiseless and missing samples under the same framework of optimal estimation. Initially, DT-CWT is used to decompose an input low-resolution noisy image into low and high frequency subbands. The high-frequency subband images are interpolated by linear minimum mean square estimation (LMMSE) based interpolation, which preserves the edges of the interpolated images. For each noisy LR image sample, we compute multiple estimates of it along different directions and then fuse those directional estimates for a more accurate denoised LR image. The estimation parameters calculated in the denoising processing can be readily used to interpolate the missing samples. The inverse DT-CWT is applied on the denoised input and interpolated high frequency subband images to obtain the high resolution image. Compared with the conventional schemes that perform denoising and interpolation in tandem, the proposed DT-CWT based noisy image interpolation method can reduce many noise-caused interpolation artifacts and preserve well the image edge structures. The visual and quantitative results show that the proposed technique outperforms many of the existing denoising and interpolation methods.
Abstract: Feature-based registration is an effective technique for clinical use, because it can greatly reduce computational costs. However, this technique, which estimates the transformation by using feature points extracted from two images, may cause misalignments. To handle with this limitation, we propose to extract the salient edges and extracted control points (CP) of medical images by using efficiency of multiresolution representation of data nonsubsampled contourlet transform (NSCT) that finds the best feature points. The MR images were first decomposed using the NSCT, and then Edge and CP were extracted from bandpass directional subband of NSCT coefficients and some proposed rules. After edge and CP extraction, mutual information was adopted for the registration of feature points and translation parameters are calculated by using particle swarm optimization (PSO). The experimental results showed that the proposed method produces totally accurate performance for registration medical CT-MR images.