Abstract: Image segmentation is an important step in image
processing. Major developments in medical imaging allow
physicians to use potent and non-invasive methods in order to
evaluate structures, performance and to diagnose human diseases. In
this study, an active contour was used to extract vessel networks
from color retina images. Automatic analysis of retina vessels
facilitates calculation of arterial index which is required to diagnose
some certain retinopathies.
Abstract: In order to explore the relationship of promotion activities, destination attribute and destination image of Vietnam and find possible solutions, this study uses decision system analysis (DSA) method to develop flowcharts based on three rounds of expert interviews. The interviews were conducted with the experts who were confirmed to directly participate or influence on the decision making that drives the promotion of Vietnam tourism process. This study identifies three models and describes specific decisions on promotion activities, destination attributes and destination images. This study finally derives a general model for promoting the Tourism Industrial Service Network (TISN) in Vietnam. This study finds that the coordination with all sectors and industries of tourism to facilitate favorable condition and improving destination attributes in linking with the efficient promotion activities is highly recommended in order to make visitors satisfied and improve the destination image.
Abstract: Mammographic images and data analysis to
facilitate modelling or computer aided diagnostic (CAD) software development should best be done using a common database that can handle various mammographic image file
formats and relate these to other patient information.
This would optimize the use of the data as both primary
reporting and enhanced information extraction of research data could be performed from the single dataset. One desired
improvement is the integration of DICOM file header information into the database, as an efficient and reliable source of supplementary patient information intrinsically
available in the images.
The purpose of this paper was to design a suitable database to link and integrate different types of image files and gather common information that can be further used for research
purposes. An interface was developed for accessing, adding,
updating, modifying and extracting data from the common
database, enhancing the future possible application of the data in CAD processing.
Technically, future developments envisaged include the creation of an advanced search function to selects image files
based on descriptor combinations. Results can be further used for specific CAD processing and other research. Design of a
user friendly configuration utility for importing of the required fields from the DICOM files must be done.
Abstract: The general idea behind the filter is to average a pixel
using other pixel values from its neighborhood, but simultaneously to
take care of important image structures such as edges. The main
concern of the proposed filter is to distinguish between any variations
of the captured digital image due to noise and due to image structure.
The edges give the image the appearance depth and sharpness. A
loss of edges makes the image appear blurred or unfocused.
However, noise smoothing and edge enhancement are traditionally
conflicting tasks. Since most noise filtering behaves like a low pass
filter, the blurring of edges and loss of detail seems a natural
consequence. Techniques to remedy this inherent conflict often
encompass generation of new noise due to enhancement.
In this work a new fuzzy filter is presented for the noise reduction
of images corrupted with additive noise. The filter consists of three
stages. (1) Define fuzzy sets in the input space to computes a fuzzy
derivative for eight different directions (2) construct a set of IFTHEN
rules by to perform fuzzy smoothing according to
contributions of neighboring pixel values and (3) define fuzzy sets in
the output space to get the filtered and edged image.
Experimental results are obtained to show the feasibility of the
proposed approach with two dimensional objects.
Abstract: The purpose of this study is to introduce a new
interface program to calculate a dose distribution with Monte Carlo method in complex heterogeneous systems such as organs or tissues
in proton therapy. This interface program was developed under
MATLAB software and includes a friendly graphical user interface
with several tools such as image properties adjustment or results display. Quadtree decomposition technique was used as an image
segmentation algorithm to create optimum geometries from Computed Tomography (CT) images for dose calculations of proton
beam. The result of the mentioned technique is a number of nonoverlapped
squares with different sizes in every image. By this way
the resolution of image segmentation is high enough in and near
heterogeneous areas to preserve the precision of dose calculations
and is low enough in homogeneous areas to reduce the number of
cells directly. Furthermore a cell reduction algorithm can be used to combine neighboring cells with the same material. The validation of this method has been done in two ways; first, in comparison with experimental data obtained with 80 MeV proton beam in Cyclotron
and Radioisotope Center (CYRIC) in Tohoku University and second, in comparison with data based on polybinary tissue calibration method, performed in CYRIC. These results are presented in this paper. This program can read the output file of Monte Carlo code while region of interest is selected manually, and give a plot of dose distribution of proton beam superimposed onto the CT images.
Abstract: The purpose of this paper is to detect human in images.
This paper proposes a method for extracting human body feature descriptors consisting of projected edge component series. The feature descriptor can express appearances and shapes of human with local
and global distribution of edges. Our method evaluated with a linear SVM classifier on Daimler-Chrysler pedestrian dataset, and test with
various sub-region size. The result shows that the accuracy level of
proposed method similar to Histogram of Oriented Gradients(HOG)
feature descriptor and feature extraction process is simple and faster than existing methods.
Abstract: Fast depth estimation from binocular vision is often
desired for autonomous vehicles, but, most algorithms could not easily
be put into practice because of the much time cost. We present an
image-processing technique that can fast estimate depth image from
binocular vision images. By finding out the lines which present the
best matched area in the disparity space image, the depth can be
estimated. When detecting these lines, an edge-emphasizing filter is
used. The final depth estimation will be presented after the smooth
filter. Our method is a compromise between local methods and global
optimization.
Abstract: In recent years, IT convergence technology has been developed to get creative solution by combining robotics or sports science technology. Object detection and recognition have mainly applied to sports science field that has processed by recognizing face and by tracking human body. But object detection and recognition using vision sensor is challenge task in real world because of illumination. In this paper, object detection and recognition using vision sensor applied to sports simulator has been introduced. Face recognition has been processed to identify user and to update automatically a person athletic recording. Human body has tracked to offer a most accurate way of riding horse simulator. Combined image processing has been processed to reduce illumination adverse affect because illumination has caused low performance in detection and recognition in real world application filed. Face has recognized using standard face graph and human body has tracked using pose model, which has composed of feature nodes generated diverse face and pose images. Face recognition using Gabor wavelet and pose recognition using pose graph is robust to real application. We have simulated using ETRI database, which has constructed on horse riding simulator.
Abstract: While compressing text files is useful, compressing
still image files is almost a necessity. A typical image takes up much
more storage than a typical text message and without compression
images would be extremely clumsy to store and distribute. The
amount of information required to store pictures on modern
computers is quite large in relation to the amount of bandwidth
commonly available to transmit them over the Internet and
applications. Image compression addresses the problem of reducing
the amount of data required to represent a digital image. Performance
of any image compression method can be evaluated by measuring the
root-mean-square-error & peak signal to noise ratio. The method of
image compression that will be analyzed in this paper is based on the
lossy JPEG image compression technique, the most popular
compression technique for color images. JPEG compression is able to
greatly reduce file size with minimal image degradation by throwing
away the least “important" information. In JPEG, both color
components are downsampled simultaneously, but in this paper we
will compare the results when the compression is done by
downsampling the single chroma part. In this paper we will
demonstrate more compression ratio is achieved when the
chrominance blue is downsampled as compared to downsampling the
chrominance red in JPEG compression. But the peak signal to noise
ratio is more when the chrominance red is downsampled as compared
to downsampling the chrominance blue in JPEG compression. In
particular we will use the hats.jpg as a demonstration of JPEG
compression using low pass filter and demonstrate that the image is
compressed with barely any visual differences with both methods.
Abstract: This paper proposes a hybrid method for eyes localization
in facial images. The novelty is in combining techniques
that utilise colour, edge and illumination cues to improve accuracy.
The method is based on the observation that eye regions have dark
colour, high density of edges and low illumination as compared
to other parts of face. The first step in the method is to extract
connected regions from facial images using colour, edge density and
illumination cues separately. Some of the regions are then removed
by applying rules that are based on the general geometry and shape
of eyes. The remaining connected regions obtained through these
three cues are then combined in a systematic way to enhance the
identification of the candidate regions for the eyes. The geometry
and shape based rules are then applied again to further remove the
false eye regions. The proposed method was tested using images from
the PICS facial images database. The proposed method has 93.7%
and 87% accuracies for initial blobs extraction and final eye detection
respectively.
Abstract: A mammography image is composed of low contrast area where the breast tissues and the breast abnormalities such as microcalcification can hardly be differentiated by the medical practitioner. This paper presents the application of active contour models (Snakes) for the segmentation of microcalcification in mammography images. Comparison on the microcalcifiation areas segmented by the Balloon Snake, Gradient Vector Flow (GVF) Snake, and Distance Snake is done against the true value of the microcalcification area. The true area value is the average microcalcification area in the original mammography image traced by the expert radiologists. From fifty images tested, the result obtained shows that the accuracy of the Balloon Snake, GVF Snake, and Distance Snake in segmenting boundaries of microcalcification are 96.01%, 95.74%, and 95.70% accuracy respectively. This implies that the Balloon Snake is a better segmentation method to locate the exact boundary of a microcalcification region.
Abstract: The algorithm represents the DCT coefficients to concentrate signal energy and proposes combination and dictator to eliminate the correlation in the same level subband for encoding the DCT-based images. This work adopts DCT and modifies the SPIHT algorithm to encode DCT coefficients. The proposed algorithm also provides the enhancement function in low bit rate in order to improve the perceptual quality. Experimental results indicate that the proposed technique improves the quality of the reconstructed image in terms of both PSNR and the perceptual results close to JPEG2000 at the same bit rate.
Abstract: This study aimed at developing visualization tools for integrating CloudSat images and Water Vapor Satellite images. KML was used for integrating data from CloudSat Satellite and GMS-6 Water Vapor Satellite. CloudSat 2D images were transformed into 3D polygons in order to achieve 3D images. Before overlaying the images on Google Earth, GMS-6 water vapor satellite images had to be rescaled into linear images. Web service was developed using webMathematica. Shoreline from GMS-6 images was compared with shoreline from LandSat images on Google Earth for evaluation. The results showed that shoreline from GMS-6 images was highly matched with the shoreline in LandSat images from Google Earth. For CloudSat images, the visualizations were compared with GMS-6 images on Google Earth. The results showed that CloudSat and GMS-6 images were highly correlated.
Abstract: Crucial information barely visible to the human eye is
often embedded in a series of low resolution images taken of the
same scene. Super resolution reconstruction is the process of
combining several low resolution images into a single higher
resolution image. The ideal algorithm should be fast, and should add
sharpness and details, both at edges and in regions without adding
artifacts. In this paper we propose a super resolution blind
reconstruction technique for linearly degraded images. In our
proposed technique the algorithm is divided into three parts an image
registration, wavelets based fusion and an image restoration. In this
paper three low resolution images are considered which may sub
pixels shifted, rotated, blurred or noisy, the sub pixel shifted images
are registered using affine transformation model; A wavelet based
fusion is performed and the noise is removed using soft thresolding.
Our proposed technique reduces blocking artifacts and also
smoothens the edges and it is also able to restore high frequency
details in an image. Our technique is efficient and computationally
fast having clear perspective of real time implementation.
Abstract: An end-member selection method for spectral unmixing that is based on Particle Swarm Optimization (PSO) is developed in this paper. The algorithm uses the K-means clustering algorithm and a method of dynamic selection of end-members subsets to find the appropriate set of end-members for a given set of multispectral images. The proposed algorithm has been successfully applied to test image sets from various platforms such as LANDSAT 5 MSS and NOAA's AVHRR. The experimental results of the proposed algorithm are encouraging. The influence of different values of the algorithm control parameters on performance is studied. Furthermore, the performance of different versions of PSO is also investigated.
Abstract: This paper presents the region based segmentation method for ultrasound images using local statistics. In this segmentation approach the homogeneous regions depends on the image granularity features, where the interested structures with dimensions comparable to the speckle size are to be extracted. This method uses a look up table comprising of the local statistics of every pixel, which are consisting of the homogeneity and similarity bounds according to the kernel size. The shape and size of the growing regions depend on this look up table entries. The algorithms are implemented by using connected seeded region growing procedure where each pixel is taken as seed point. The region merging after the region growing also suppresses the high frequency artifacts. The updated merged regions produce the output in formed of segmented image. This algorithm produces the results that are less sensitive to the pixel location and it also allows a segmentation of the accurate homogeneous regions.
Abstract: We present a method for fast volume rendering using
graphics hardware (GPU). To our knowledge, it is the first implementation
on the GPU. Based on the Shear-Warp algorithm, our
GPU-based method provides real-time frame rates and outperforms
the CPU-based implementation. When the number of slices is not
sufficient, we add in-between slices computed by interpolation. This
improves then the quality of the rendered images. We have also
implemented the ray marching algorithm on the GPU. The results
generated by the three algorithms (CPU-based and GPU-based Shear-
Warp, GPU-based Ray Marching) for two test models has proved that
the ray marching algorithm outperforms the shear-warp methods in
terms of speed up and image quality.
Abstract: In this paper we describe a computer-aided diagnosis (CAD) system for automated detection of pulmonary nodules in computed-tomography (CT) images. After extracting the pulmonary parenchyma using a combination of image processing techniques, a region growing method is applied to detect nodules based on 3D geometric features. We applied the CAD system to CT scans collected in a screening program for lung cancer detection. Each scan consists of a sequence of about 300 slices stored in DICOM (Digital Imaging and Communications in Medicine) format. All malignant nodules were detected and a low false-positive detection rate was achieved.
Abstract: In quality control of freeze-dried durian, crispiness is
a key quality index of the product. Generally, crispy testing has to be
done by a destructive method. A nondestructive testing of the
crispiness is required because the samples can be reused for other
kinds of testing. This paper proposed a crispiness classification
method of freeze-dried durians using fuzzy logic for decision
making. The physical changes of a freeze-dried durian include the
pores appearing in the images. Three physical features including (1)
the diameters of pores, (2) the ratio of the pore area and the
remaining area, and (3) the distribution of the pores are considered to
contribute to the crispiness. The fuzzy logic is applied for making the
decision. The experimental results comparing with food expert
opinion showed that the accuracy of the proposed classification
method is 83.33 percent.
Abstract: For over a decade, the Pulse Coupled Neural Network
(PCNN) based algorithms have been successfully used in image
interpretation applications including image segmentation. There are
several versions of the PCNN based image segmentation methods,
and the segmentation accuracy of all of them is very sensitive to the
values of the network parameters. Most methods treat PCNN
parameters like linking coefficient and primary firing threshold as
global parameters, and determine them by trial-and-error. The
automatic determination of appropriate values for linking coefficient,
and primary firing threshold is a challenging problem and deserves
further research. This paper presents a method for obtaining global as
well as local values for the linking coefficient and the primary firing
threshold for neurons directly from the image statistics. Extensive
simulation results show that the proposed approach achieves
excellent segmentation accuracy comparable to the best accuracy
obtainable by trial-and-error for a variety of images.