Abstract: The main goal of the present work is to decrease the
computational burden for optimum design of steel frames with
frequency constraints using a new type of neural networks called
Wavelet Neural Network. It is contested to train a suitable neural
network for frequency approximation work as the analysis program.
The combination of wavelet theory and Neural Networks (NN)
has lead to the development of wavelet neural networks.
Wavelet neural networks are feed-forward networks using
wavelet as activation function. Wavelets are mathematical
functions within suitable inner parameters, which help them to
approximate arbitrary functions. WNN was used to predict the
frequency of the structures. In WNN a RAtional function with
Second order Poles (RASP) wavelet was used as a transfer
function. It is shown that the convergence speed was faster
than other neural networks. Also comparisons of WNN with
the embedded Artificial Neural Network (ANN) and with
approximate techniques and also with analytical solutions are
available in the literature.
Abstract: The purpose of this research is to develop a security model for voice eavesdropping protection over digital networks. The proposed model provides an encryption scheme and a personal secret key exchange between communicating parties, a so-called voice data transformation system, resulting in a real-privacy conversation. The operation of this system comprises two main steps as follows: The first one is the personal secret key exchange for using the keys in the data encryption process during conversation. The key owner could freely make his/her choice in key selection, so it is recommended that one should exchange a different key for a different conversational party, and record the key for each case into the memory provided in the client device. The next step is to set and record another personal option of encryption, either taking all frames or just partial frames, so-called the figure of 1:M. Using different personal secret keys and different sets of 1:M to different parties without the intervention of the service operator, would result in posing quite a big problem for any eavesdroppers who attempt to discover the key used during the conversation, especially in a short period of time. Thus, it is quite safe and effective to protect the case of voice eavesdropping. The results of the implementation indicate that the system can perform its function accurately as designed. In this regard, the proposed system is suitable for effective use in voice eavesdropping protection over digital networks, without any requirements to change presently existing network systems, mobile phone network and VoIP, for instance.
Abstract: In this paper, a new reversible watermarking method is presented that reduces the size of a stereoscopic image sequence while keeping its content visible. The proposed technique embeds the residuals of the right frames to the corresponding frames of the left sequence, halving the total capacity. The residual frames may result in after a disparity compensated procedure between the two video streams or by a joint motion and disparity compensation. The residuals are usually lossy compressed before embedding because of the limited embedding capacity of the left frames. The watermarked frames are visible at a high quality and at any instant the stereoscopic video may be recovered by an inverse process. In fact, the left frames may be exactly recovered whereas the right ones are slightly distorted as the residuals are not embedded intact. The employed embedding method reorders the left frame into an array of consecutive pixel pairs and embeds a number of bits according to their intensity difference. In this way, it hides a number of bits in intensity smooth areas and most of the data in textured areas where resulting distortions are less visible. The experimental evaluation demonstrates that the proposed scheme is quite effective.
Abstract: In this paper we present a novel method, which
reduces the computational complexity of abrupt cut detection. We
have proposed fast algorithm, where the similarity of frames within
defined step is evaluated instead of comparing successive frames.
Based on the results of simulation on large video collection, the
proposed fast algorithm is able to achieve 80% reduction of needed
frames comparisons compared to actually used methods without the
shot cut detection accuracy degradation.
Abstract: Modeling of Panel Zone (PZ) seismic behavior,
because of its role in overall ductility and lateral stiffness of steel
moment frames, has been considered a challenge for years. There are
some studies regarding the effects of different doubler plates
thicknesses and geometric properties of PZ on its seismic behavior.
However, there is not much investigation on the effects of number of
provided continuity plates in case of presence of one triangular
haunch, two triangular haunches and rectangular haunch (T shape
haunches) for exterior columns. In this research first detailed finite
element models of 12tested connection of SAC joint venture were
created and analyzed then obtained cyclic behavior backbone curves
of these models besides other FE models for similar tests were used
for neural network training. Then seismic behavior of these data is
categorized according to continuity plate-s arrangements and
differences in type of haunches. PZ with one-sided haunches have
little plastic rotation. As the number of continuity plates increases
due to presence of two triangular haunches (four continuity plate),
there will be no plastic rotation, in other words PZ behaves in its
elastic range. In the case of rectangular haunch, PZ show more plastic
rotation in comparison with one-sided triangular haunch and
especially double-sided triangular haunches. Moreover, the models
that will be presented in case of triangular one-sided and double-
sided haunches and rectangular haunches as a result of this study
seem to have a proper estimation of PZ seismic behavior.
Abstract: Chevron frames (Inverted-V-braced frames or Vbraced
frames) have seismic disadvantages, such as not good exhibit force redistribution capability and compression brace buckles
immediately. Researchers developed new design provisions on
increasing both the ductility and lateral resistance of these structures
in seismic areas. One of these new methods is adding zipper columns, as proposed by Khatib et al. (1988) [2]. Zipper columns are
vertical members connecting the intersection points of the braces
above the first floor. In this paper applicability of the suspended
zipper system to Seismic Rehabilitation of Steel Structures is investigated.
The models are 3-, 6-, 9-, and 12-story Inverted-V-braced frames.
In this case, it is assumed that the structures must be rehabilitated. For rehabilitation of structures, zipper column is used. The result of
researches showed that the suspended zipper system is effective in
case of 3-, 6-, and 9-story Inverted-V-braced frames and it would
increase lateral resistance of structure up to life safety level. But in
case of high-rise buildings (such as 12 story frame), it doesn-t show
good performance. For solving this problem, the braced bay can
consist of small “units" over the height of the entire structure, which each of them is a zipper-braced bay with a few stories. By using this
method the lateral resistance of 12 story Inverted-V-braced frames is increased up to safety life level.
Abstract: In this paper we proposed a method for finding video
frames representing one sign in the finger alphabet. The method is
based on determining hands location, segmentation and the use of
standard video quality evaluation metrics. Metric calculation is
performed only in regions of interest. Sliding mechanism for finding
local extrema and adaptive threshold based on local averaging is used
for key frames selection. The success rate is evaluated by recall,
precision and F1 measure. The method effectiveness is compared
with metrics applied to all frames. Proposed method is fast, effective
and relatively easy to realize by simple input video preprocessing
and subsequent use of tools designed for video quality measuring.
Abstract: In this paper we present a method of abrupt cut detection with a novel logic of frames- comparison. Actual frame is compared with its motion estimated prediction instead of comparison with successive frame. Four different similarity metrics were employed to estimate the resemblance of compared frames. Obtained results were evaluated by standard used measures of test accuracy and compared with existing approach. Based on the results, we claim the proposed method is more effective and Pearson correlation coefficient obtained the best results among chosen similarity metrics.
Abstract: We study dynamic instability in high-rise steel moment
resisting frames (SMRFs) subjected to synthetic long-period ground
motions caused by hypothetical huge subduction earthquakes. Since
long duration as well as long dominant periods is a characteristic of
long-period ground motions, interstory drifts may enter the negative
postyield stiffness range many times when high-rise buildings are
subjected to long-period ground motions. Through the case studies of
9 high-rise SMRFs designed in accordance with the Japanese design
practice in 1980s, we demonstrate that drifting, or accumulation of
interstory drifts in one direction, occurs at the lower stories of the
SMRFs, if their natural periods are close to the dominant periods of the
long-period ground motions. The drifting led to residual interstory
drift ratio over 0.01, or to collapse if the design base shear was small.
Abstract: Image processing for capsule endoscopy requires large
memory and it takes hours for diagnosis since operation time is
normally more than 8 hours. A real-time analysis algorithm of capsule
images can be clinically very useful. It can differentiate abnormal
tissue from health structure and provide with correlation information
among the images. Bleeding is our interest in this regard and we
propose a method of detecting frames with potential bleeding in
real-time. Our detection algorithm is based on statistical analysis and
the shapes of bleeding spots. We tested our algorithm with 30 cases of
capsule endoscopy in the digestive track. Results were excellent where
a sensitivity of 99% and a specificity of 97% were achieved in
detecting the image frames with bleeding spots.
Abstract: In this paper, a fast motion compensation algorithm is
proposed that improves coding efficiency for video sequences with
brightness variations. We also propose a cross entropy measure
between histograms of two frames to detect brightness variations. The
framewise brightness variation parameters, a multiplier and an offset
field for image intensity, are estimated and compensated. Simulation
results show that the proposed method yields a higher peak signal to
noise ratio (PSNR) compared with the conventional method, with a
greatly reduced computational load, when the video scene contains
illumination changes.
Abstract: The paper presents an on-line recognition machine
(RM) for continuous/isolated, dynamic and static gestures that arise
in Flight Deck Officer (FDO) training. RM is based on generic pattern
recognition framework. Gestures are represented as templates using
summary statistics. The proposed recognition algorithm exploits temporal
and spatial characteristics of gestures via dynamic programming
and Markovian process. The algorithm predicts corresponding index
of incremental input data in the templates in an on-line mode.
Accumulated consistency in the sequence of prediction provides a
similarity measurement (Score) between input data and the templates.
The algorithm provides an intuitive mechanism for automatic detection
of start/end frames of continuous gestures. In the present paper,
we consider isolated gestures. The performance of RM is evaluated
using four datasets - artificial (W TTest), hand motion (Yang) and
FDO (tracker, vision-based ). RM achieves comparable results which
are in agreement with other on-line and off-line algorithms such as
hidden Markov model (HMM) and dynamic time warping (DTW).
The proposed algorithm has the additional advantage of providing
timely feedback for training purposes.
Abstract: As the development of digital technology is increasing,
Digital cinema is getting more spread.
However, content copy and attack against the digital cinema becomes
a serious problem. To solve the above security problem, we propose
“Additional Watermarking" for digital cinema delivery system. With
this proposed “Additional watermarking" method, we protect content
copyrights at encoder and user side information at decoder. It realizes
the traceability of the watermark embedded at encoder.
The watermark is embedded into the random-selected frames using
Hash function. Using it, the embedding position is distributed by Hash
Function so that third parties do not break off the watermarking
algorithm.
Finally, our experimental results show that proposed method is much
better than the convenient watermarking techniques in terms of
robustness, image quality and its simple but unbreakable algorithm.
Abstract: Video watermarking is usually considered as watermarking of a set of still images. In frame-by-frame watermarking approach, each video frame is seen as a single watermarked image, so collusion attack is more critical in video watermarking. If the same or redundant watermark is used for embedding in every frame of video, the watermark can be estimated and then removed by watermark estimate remodolulation (WER) attack. Also if uncorrelated watermarks are used for every frame, these watermarks can be washed out with frame temporal filtering (FTF). Switching watermark system or so-called SS-N system has better performance against WER and FTF attacks. In this system, for each frame, the watermark is randomly picked up from a finite pool of watermark patterns. At first SS-N system will be surveyed and then a new collusion attack for SS-N system will be proposed using a new algorithm for separating video frame based on watermark pattern. So N sets will be built in which every set contains frames carrying the same watermark. After that, using WER attack in every set, N different watermark patterns will be estimated and removed later.
Abstract: Sign language is used by the deaf and hard of hearing people for communication. Automatic sign language recognition is a challenging research area since sign language often is the only way of communication for the deaf people. Sign language includes different components of visual actions made by the signer using the hands, the face, and the torso, to convey his/her meaning. To use different aspects of signs, we combine the different groups of features which have been extracted from the image frames recorded directly by a stationary camera. We combine the features in two levels by employing three techniques. At the feature level, an early feature combination can be performed by concatenating and weighting different feature groups, or by concatenating feature groups over time and using LDA to choose the most discriminant elements. At the model level, a late fusion of differently trained models can be carried out by a log-linear model combination. In this paper, we investigate these three combination techniques in an automatic sign language recognition system and show that the recognition rate can be significantly improved.
Abstract: Video Mosaicing is the stitching of selected frames of
a video by estimating the camera motion between the frames and
thereby registering successive frames of the video to arrive at the
mosaic. Different techniques have been proposed in the literature for
video mosaicing. Despite of the large number of papers dealing with
techniques to generate mosaic, only a few authors have investigated
conditions under which these techniques generate good estimate of
motion parameters. In this paper, these techniques are studied under
different videos, and the reasons for failures are found. We propose
algorithms with incorporation of outlier removal algorithms for better
estimation of motion parameters.
Abstract: Vision based tracking problem is solved through a
combination of optical flow, MACH filter and log r-θ mapping.
Optical flow is used for detecting regions of movement in video
frames acquired under variable lighting conditions. The region of
movement is segmented and then searched for the target. A template
is used for target recognition on the segmented regions for detecting
the region of interest. The template is trained offline on a sequence of
target images that are created using the MACH filter and log r-θ
mapping. The template is applied on areas of movement in
successive frames and strong correlation is seen for in-class targets.
Correlation peaks above a certain threshold indicate the presence of
target and the target is tracked over successive frames.
Abstract: In this paper, we propose a novel adaptive
spatiotemporal filter that utilizes image sequences in order to remove
noise. The consecutive frames include: current, previous and next
noisy frames. The filter proposed in this paper is based upon the
weighted averaging pixels intensity and noise variance in image
sequences. It utilizes the Appropriate Number of Consecutive Frames
(ANCF) based on the noisy pixels intensity among the frames. The
number of consecutive frames is adaptively calculated for each
region in image and its value may change from one region to another
region depending on the pixels intensity within the region. The
weights are determined by a well-defined mathematical criterion,
which is adaptive to the feature of spatiotemporal pixels of the
consecutive frames. It is experimentally shown that the proposed
filter can preserve image structures and edges under motion while
suppressing noise, and thus can be effectively used in image
sequences filtering. In addition, the AWA filter using ANCF is
particularly well suited for filtering sequences that contain segments
with abruptly changing scene content due to, for example, rapid
zooming and changes in the view of the camera.
Abstract: This paper describes the crashworthiness assessment and improvement of tlting train made of sandwich composites. The crashworhiness assessment of tilting train was conducted according to four collision scenarios of the Korean railway safety law. Collision analysis was carried out using explicit finite element analysis code LS-DYNA 3D. The finite element model consists of 3-D finite element model and 1-D equivalent model to save the finite element modeling and calculation time. It found that the crashworthiness analysis results were satisfied with the performance requirements except the crash scenario-2. In order to meet the crashworthiness requirements for crash scenario-2, the stiffness reinforcement for the laminate composite cover and metal frames of cabmask structure were proposed. Consequentially, it has satisfied the requirement for crash scenario-2.
Abstract: In this paper we present a new method for over-height
vehicle detection in low headroom streets and highways using digital
video possessing. The accuracy and the lower price comparing to
present detectors like laser radars and the capability of providing
extra information like speed and height measurement make this
method more reliable and efficient. In this algorithm the features are
selected and tracked using KLT algorithm. A blob extraction
algorithm is also applied using background estimation and
subtraction. Then the world coordinates of features that are inside the
blobs are estimated using a noble calibration method. As, the heights
of the features are calculated, we apply a threshold to select overheight
features and eliminate others. The over-height features are
segmented using some association criteria and grouped using an
undirected graph. Then they are tracked through sequential frames.
The obtained groups refer to over-height vehicles in a scene.