Abstract: In this paper we propose a novel method for human
face segmentation using the elliptical structure of the human head. It
makes use of the information present in the edge map of the image.
In this approach we use the fact that the eigenvalues of covariance
matrix represent the elliptical structure. The large and small
eigenvalues of covariance matrix are associated with major and
minor axial lengths of an ellipse. The other elliptical parameters are
used to identify the centre and orientation of the face. Since an
Elliptical Hough Transform requires 5D Hough Space, the Circular
Hough Transform (CHT) is used to evaluate the elliptical parameters.
Sparse matrix technique is used to perform CHT, as it squeeze zero
elements, and have only a small number of non-zero elements,
thereby having an advantage of less storage space and computational
time. Neighborhood suppression scheme is used to identify the valid
Hough peaks. The accurate position of the circumference pixels for
occluded and distorted ellipses is identified using Bresenham-s
Raster Scan Algorithm which uses the geometrical symmetry
properties. This method does not require the evaluation of tangents
for curvature contours, which are very sensitive to noise. The method
has been evaluated on several images with different face orientations.
Abstract: Purpose: To develop a method for automatic segmentation of adipose and muscular tissue in thighs from magnetic resonance images. Materials and methods: Thirty obese women were scanned on a Siemens Impact Expert 1T resonance machine. 1500 images were finally used in the tests. The developed segmentation method is a recursive and multilevel process that makes use of several concepts such as shaped histograms, adaptative thresholding and connectivity. The segmentation process was implemented in Matlab and operates without the need of any user interaction. The whole set of images were segmented with the developed method. An expert radiologist segmented the same set of images following a manual procedure with the aid of the SliceOmatic software (Tomovision). These constituted our 'goal standard'. Results: The number of coincidental pixels of the automatic and manual segmentation procedures was measured. The average results were above 90 % of success in most of the images. Conclusions: The proposed approach allows effective automatic segmentation of MRIs from thighs, comparable to expert manual performance.
Abstract: A novel feature selection strategy to improve the recognition accuracy on the faces that are affected due to nonuniform illumination, partial occlusions and varying expressions is proposed in this paper. This technique is applicable especially in scenarios where the possibility of obtaining a reliable intra-class probability distribution is minimal due to fewer numbers of training samples. Phase congruency features in an image are defined as the points where the Fourier components of that image are maximally inphase. These features are invariant to brightness and contrast of the image under consideration. This property allows to achieve the goal of lighting invariant face recognition. Phase congruency maps of the training samples are generated and a novel modular feature selection strategy is implemented. Smaller sub regions from a predefined neighborhood within the phase congruency images of the training samples are merged to obtain a large set of features. These features are arranged in the order of increasing distance between the sub regions involved in merging. The assumption behind the proposed implementation of the region merging and arrangement strategy is that, local dependencies among the pixels are more important than global dependencies. The obtained feature sets are then arranged in the decreasing order of discriminating capability using a criterion function, which is the ratio of the between class variance to the within class variance of the sample set, in the PCA domain. The results indicate high improvement in the classification performance compared to baseline algorithms.
Abstract: Skin color can provide a useful and robust cue
for human-related image analysis, such as face detection,
pornographic image filtering, hand detection and tracking,
people retrieval in databases and Internet, etc. The major
problem of such kinds of skin color detection algorithms is
that it is time consuming and hence cannot be applied to a real
time system. To overcome this problem, we introduce a new
fast technique for skin detection which can be applied in a real
time system. In this technique, instead of testing each image
pixel to label it as skin or non-skin (as in classic techniques),
we skip a set of pixels. The reason of the skipping process is
the high probability that neighbors of the skin color pixels are
also skin pixels, especially in adult images and vise versa. The
proposed method can rapidly detect skin and non-skin color
pixels, which in turn dramatically reduce the CPU time
required for the protection process. Since many fast detection
techniques are based on image resizing, we apply our
proposed pixel skipping technique with image resizing to
obtain better results. The performance evaluation of the
proposed skipping and hybrid techniques in terms of the
measured CPU time is presented. Experimental results
demonstrate that the proposed methods achieve better result
than the relevant classic method.
Abstract: In this study, a novel approach of image embedding is introduced. The proposed method consists of three main steps. First, the edge of the image is detected using Sobel mask filters. Second, the least significant bit LSB of each pixel is used. Finally, a gray level connectivity is applied using a fuzzy approach and the ASCII code is used for information hiding. The prior bit of the LSB represents the edged image after gray level connectivity, and the remaining six bits represent the original image with very little difference in contrast. The proposed method embeds three images in one image and includes, as a special case of data embedding, information hiding, identifying and authenticating text embedded within the digital images. Image embedding method is considered to be one of the good compression methods, in terms of reserving memory space. Moreover, information hiding within digital image can be used for security information transfer. The creation and extraction of three embedded images, and hiding text information is discussed and illustrated, in the following sections.
Abstract: In Geographic Information System, one of the sources
of obtaining needed geographic data is digitizing analog maps and
evaluation of aerial and satellite photos. In this study, a method will
be discussed which can be used to extract vectorial features and
creating vectorized drawing files for aerial photos. At the same time
a software developed for these purpose. Converting from raster to
vector is also known as vectorization and it is the most important step
when creating vectorized drawing files. In the developed algorithm,
first of all preprocessing on the aerial photo is done. These are;
converting to grayscale if necessary, reducing noise, applying some
filters and determining the edge of the objects etc. After these steps,
every pixel which constitutes the photo are followed from upper left
to right bottom by examining its neighborhood relationship and one
pixel wide lines or polylines obtained. The obtained lines have to be
erased for preventing confusion while continuing vectorization
because if not erased they can be perceived as new line, but if erased
it can cause discontinuity in vector drawing so the image converted
from 2 bit to 8 bit and the detected pixels are expressed as a different
bit. In conclusion, the aerial photo can be converted to vector form
which includes lines and polylines and can be opened in any CAD
application.
Abstract: This article presents a computationally tractable probabilistic model for the relation between the complex wavelet coefficients of two images of the same scene. The two images are acquisitioned at distinct moments of times, or from distinct viewpoints, or by distinct sensors. By means of the introduced probabilistic model, we argue that the similarity between the two images is controlled not by the values of the wavelet coefficients, which can be altered by many factors, but by the nature of the wavelet coefficients, that we model with the help of hidden state variables. We integrate this probabilistic framework in the construction of a new image registration algorithm. This algorithm has sub-pixel accuracy and is robust to noise and to other variations like local illumination changes. We present the performance of our algorithm on various image types.
Abstract: The purpose of this work is to present a method for
rigid registration of medical images using 1D binary projections
when a part of one of the two images is missing. We use 1D binary
projections and we adjust the projection limits according to the
reduced image in order to perform accurate registration. We use the
variance of the weighted ratio as a registration function which we
have shown is able to register 2D and 3D images more accurately and
robustly than mutual information methods. The function is computed
explicitly for n=5 Chebyshev points in a [-9,+9] interval and it is
approximated using Chebyshev polynomials for all other points. The
images used are MR scans of the head. We find that the method is
able to register the two images with average accuracy 0.3degrees for
rotations and 0.2 pixels for translations for a y dimension of 156 with
initial dimension 256. For y dimension 128/256 the accuracy
decreases to 0.7 degrees for rotations and 0.6 pixels for translations.
Abstract: This paper presents a comparative analysis of a new
unsupervised PCA-based technique for steel plates texture segmentation
towards defect detection. The proposed scheme called Variance
Based Component Analysis or VBCA employs PCA for feature
extraction, applies a feature reduction algorithm based on variance of
eigenpictures and classifies the pixels as defective and normal. While
the classic PCA uses a clusterer like Kmeans for pixel clustering,
VBCA employs thresholding and some post processing operations to
label pixels as defective and normal. The experimental results show
that proposed algorithm called VBCA is 12.46% more accurate and
78.85% faster than the classic PCA.
Abstract: The aim of this research is to develop a fast and
reliable surveillance system based on a personal digital assistant
(PDA) device. This is to extend the capability of the device to detect
moving objects which is already available in personal computers.
Secondly, to compare the performance between Background
subtraction (BS) and Temporal Frame Differencing (TFD) techniques
for PDA platform as to which is more suitable. In order to reduce
noise and to prepare frames for the moving object detection part,
each frame is first converted to a gray-scale representation and then
smoothed using a Gaussian low pass filter. Two moving object
detection schemes i.e., BS and TFD have been analyzed. The
background frame is updated by using Infinite Impulse Response
(IIR) filter so that the background frame is adapted to the varying
illuminate conditions and geometry settings. In order to reduce the
effect of noise pixels resulting from frame differencing
morphological filters erosion and dilation are applied. In this
research, it has been found that TFD technique is more suitable for
motion detection purpose than the BS in term of speed. On average
TFD is approximately 170 ms faster than the BS technique
Abstract: Iris pattern is an important biological feature of human body; it becomes very hot topic in both research and practical applications. In this paper, an algorithm is proposed for iris recognition and a simple, efficient and fast method is introduced to extract a set of discriminatory features using first order gradient operator applied on grayscale images. The gradient based features are robust, up to certain extents, against the variations may occur in contrast or brightness of iris image samples; the variations are mostly occur due lightening differences and camera changes. At first, the iris region is located, after that it is remapped to a rectangular area of size 360x60 pixels. Also, a new method is proposed for detecting eyelash and eyelid points; it depends on making image statistical analysis, to mark the eyelash and eyelid as a noise points. In order to cover the features localization (variation), the rectangular iris image is partitioned into N overlapped sub-images (blocks); then from each block a set of different average directional gradient densities values is calculated to be used as texture features vector. The applied gradient operators are taken along the horizontal, vertical and diagonal directions. The low order norms of gradient components were used to establish the feature vector. Euclidean distance based classifier was used as a matching metric for determining the degree of similarity between the features vector extracted from the tested iris image and template features vectors stored in the database. Experimental tests were performed using 2639 iris images from CASIA V4-Interival database, the attained recognition accuracy has reached up to 99.92%.
Abstract: A higher order spline interpolated contour obtained
with up-sampling of homogenously distributed coordinates for
segmentation of kidney region in different classes of ultrasound
kidney images has been developed and presented in this paper. The
performance of the proposed method is measured and compared with
modified snake model contour, Markov random field contour and
expert outlined contour. The validation of the method is made in
correspondence with expert outlined contour using maximum coordinate
distance, Hausdorff distance and mean radial distance
metrics. The results obtained reveal that proposed scheme provides
optimum contour that agrees well with expert outlined contour.
Moreover this technique helps to preserve the pixels-of-interest
which in specific defines the functional characteristic of kidney. This
explores various possibilities in implementing computer-aided
diagnosis system exclusively for US kidney images.
Abstract: In this paper three basic approaches and different
methods under each of them for extracting region of interest (ROI)
from stationary images are explored. The results obtained for each of
the proposed methods are shown, and it is demonstrated where each
method outperforms the other. Two main problems in ROI
extraction: the channel selection problem and the saliency reversal
problem are discussed and how best these two are addressed by
various methods is also seen. The basic approaches are 1) Saliency
based approach 2) Wavelet based approach 3) Clustering based
approach. The saliency approach performs well on images containing
objects of high saturation and brightness. The wavelet based
approach performs well on natural scene images that contain regions
of distinct textures. The mean shift clustering approach partitions the
image into regions according to the density distribution of pixel
intensities. The experimental results of various methodologies show
that each technique performs at different acceptable levels for
various types of images.
Abstract: Detection of player identity is challenging task in sport video content analysis. In case of soccer video player number recognition is effective and precise solution. Jersey numbers can be considered as scene text and difficulties in localization and recognition appear due to variations in orientation, size, illumination, motion etc. This paper proposed new method for player number localization and recognition. By observing hue, saturation and value for 50 different jersey examples we noticed that most often combination of low and high saturated pixels is used to separate number and jersey region. Image segmentation method based on this observation is introduced. Then, novel method for player number localization based on internal contours is proposed. False number candidates are filtered using area and aspect ratio. Before OCR processing extracted numbers are enhanced using image smoothing and rotation normalization.
Abstract: The approach based on the wavelet transform has
been widely used for image denoising due to its multi-resolution
nature, its ability to produce high levels of noise reduction and the
low level of distortion introduced. However, by removing noise, high
frequency components belonging to edges are also removed, which
leads to blurring the signal features. This paper proposes a new
method of image noise reduction based on local variance and edge
analysis. The analysis is performed by dividing an image into 32 x 32
pixel blocks, and transforming the data into wavelet domain. Fast
lifting wavelet spatial-frequency decomposition and reconstruction is
developed with the advantages of being computationally efficient and
boundary effects minimized. The adaptive thresholding by local
variance estimation and edge strength measurement can effectively
reduce image noise while preserve the features of the original image
corresponding to the boundaries of the objects. Experimental results
demonstrate that the method performs well for images contaminated
by natural and artificial noise, and is suitable to be adapted for
different class of images and type of noises. The proposed algorithm
provides a potential solution with parallel computation for real time
or embedded system application.
Abstract: In this work the effects of uniaxial mechanical stress on a pixel readout circuit are theoretically analyzed. It is the effects of mechanical stress on the in-pixel transistors do not arise at the output, when a correlated double sampling circuit is used. However, mechanical stress effects on the photodiode will directly appear at the readout chain output. Therefore, compensation techniques are needed to overcome this situation. Moreover simulation technique of mechanical stress is proposed and diverse layout as well as design recommendations are put forward, in order to minimize stress related effects on the output of a circuit. he shown, that wever, Moreover, a out
Abstract: One of the main limitations for the resolution of
optical instruments is the size of the sensor-s pixels. In this paper we
introduce a new sub pixel resolution algorithm to enhance the
resolution of images. This method is based on the analysis of multiimages
which are fast recorded during the fine relative motion of
image and pixel arrays of CCDs. It is shown that by applying this
method for a sample noise free image one will enhance the resolution
with 10-14 order of error.
Abstract: In this work we propose a novel Steganographic
method for hiding information within the spatial domain of the gray
scale image. The proposed approach works by dividing the cover into
blocks of equal sizes and then embeds the message in the edge of the
block depending on the number of ones in left four bits of the pixel.
The proposed approach is tested on a database consists of 100
different images. Experimental results, compared with other
methods, showed that the proposed approach hide more large
information and gave a good visual quality stego-image that can be
seen by human eyes.
Abstract: In this study, a new criterion for determining the number of classes an image should be segmented is proposed. This criterion is based on discriminant analysis for measuring the separability among the segmented classes of pixels. Based on the new discriminant criterion, two algorithms for recursively segmenting the image into determined number of classes are proposed. The proposed methods can automatically and correctly segment objects with various illuminations into separated images for further processing. Experiments on the extraction of text strings from complex document images demonstrate the effectiveness of the proposed methods.1
Abstract: This paper proposes a new approach for image encryption
using a combination of different permutation techniques.
The main idea behind the present work is that an image can be
viewed as an arrangement of bits, pixels and blocks. The intelligible
information present in an image is due to the correlations among the
bits, pixels and blocks in a given arrangement. This perceivable information
can be reduced by decreasing the correlation among the bits,
pixels and blocks using certain permutation techniques. This paper
presents an approach for a random combination of the aforementioned
permutations for image encryption. From the results, it is observed
that the permutation of bits is effective in significantly reducing the
correlation thereby decreasing the perceptual information, whereas
the permutation of pixels and blocks are good at producing higher
level security compared to bit permutation. A random combination
method employing all the three techniques thus is observed to be
useful for tactical security applications, where protection is needed
only against a casual observer.