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: 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: Rotation or tilt present in an image capture by digital
means can be detected and corrected using Artificial Neural Network
(ANN) for application with a Face Recognition System (FRS). Principal
Component Analysis (PCA) features of faces at different angles
are used to train an ANN which detects the rotation for an input image
and corrected using a set of operations implemented using another
system based on ANN. The work also deals with the recognition
of human faces with features from the foreheads, eyes, nose and
mouths as decision support entities of the system configured using
a Generalized Feed Forward Artificial Neural Network (GFFANN).
These features are combined to provide a reinforced decision for
verification of a person-s identity despite illumination variations. The
complete system performing facial image rotation detection, correction
and recognition using re-enforced decision support provides a
success rate in the higher 90s.
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: In this paper, we proposed the robust mobile object
detection method for light effect in the night street image block based
updating reference background model using block state analysis.
Experiment image is acquired sequence color video from steady
camera. When suddenly appeared artificial illumination, reference
background model update this information such as street light, sign
light. Generally natural illumination is change by temporal, but
artificial illumination is suddenly appearance. So in this paper for
exactly detect artificial illumination have 2 state process. First process
is compare difference between current image and reference
background by block based, it can know changed blocks. Second
process is difference between current image-s edge map and reference
background image-s edge map, it possible to estimate illumination at
any block. This information is possible to exactly detect object,
artificial illumination and it was generating reference background
more clearly. Block is classified by block-state analysis. Block-state
has a 4 state (i.e. transient, stationary, background, artificial
illumination). Fig. 1 is show characteristic of block-state respectively
[1]. Experimental results show that the presented approach works well
in the presence of illumination variance.
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 reports a new pattern recognition approach for face recognition. The biological model of light receptors - cones and rods in human eyes and the way they are associated with pattern vision in human vision forms the basis of this approach. The functional model is simulated using CWD and WPD. The paper also discusses the experiments performed for face recognition using the features extracted from images in the AT & T face database. Artificial Neural Network and k- Nearest Neighbour classifier algorithms are employed for the recognition purpose. A feature vector is formed for each of the face images in the database and recognition accuracies are computed and compared using the classifiers. Simulation results show that the proposed method outperforms traditional way of feature extraction methods prevailing for pattern recognition in terms of recognition accuracy for face images with pose and illumination variations.
Abstract: Indium-tin oxide films are deposited by low plasma
temperature RF sputtering on highly flexible modification of glycol
polyethyleneterephtalate substrates. The produced layers are
characterized with transparency over 82 % and sheet resistance of
86.9 Ω/square. The film’s conductivity was further improved by
additional UV illumination from light source (365 nm), having power
of 250 W. The influence of the UV exposure dose on the structural
and electro-optical properties of ITO was investigated. It was
established that the optimum time of illumination is 10 minutes and
further UV treatment leads to polymer substrates degradation.
Structural and bonds type analysis show that at longer treatment
carbon atoms release and diffuse into ITO films, which worsen their
electrical behavior. For the optimum UV dose the minimum sheet
resistance was measured to be 19.2 Ω/square, and the maximum
transparency remained almost unchanged – above 82 %.
Abstract: This paper presented a theoretical and numerical investigation of the Compact Antenna Test Range (CATR) equipped with Super Hybrid Modulated Segmented Exponential Serrations (SHMSES). The investigation was based on diffraction theory and, more specifically, the Fresnel diffraction formulation. The CATR provides uniform illumination within the Fresnel region to test antenna. Application of serrated edges has been shown to be a good method to control diffraction at the edges of the reflectors. However, in order to get some insight into the positive effect of serrated edges a less rigorous analysis technique known as Physical Optics (PO) may be used. Ripple free and enhanced quiet zone width are observed for specific values of width and height modulation factors per serrations. The performance of SHMSE serrated reflector is evaluated in order to observe the effects of edge diffraction on the test zone fields.
Abstract: Monitored 3-Dimensional (3D) video experience can be utilized as “feedback information” to fine tune the service parameters for providing a better service to the demanding 3D service customers. The 3D video experience which includes both video quality and depth perception is influenced by several contextual and content related factors (e.g., ambient illumination condition, content characteristics, etc) due to the complex nature of the 3D video. Therefore, effective factors on this experience should be utilized while assessing it. In this paper, structural information of the depth map sequences of the 3D video is considered as content related factor effective on the depth perception assessment. Cartoon-like filter is utilized to abstract the significant depth levels in the depth map sequences to determine the structural information. Moreover, subjective experiments are conducted using 3D videos associated with cartoon-like depth map sequences to investigate the effectiveness of ambient illumination condition, which is a contextual factor, on depth perception. Using the knowledge gained through this study, 3D video experience metrics can be developed to deliver better service to the 3D video service users.
Abstract: We propose photo-BJMOSFET (Bipolar Junction Metal-Oxide-Semiconductor Field Effect Transistor) fabricated on SOI film. ITO film is adopted in the device as gate electrode to reduce light absorption. I-V characteristics of photo-BJMOSFET obtained in dark (dark current) and under 570nm illumination (photo current) are studied furthermore to achieve high photo-to-dark-current contrast ratio. Two variables in the calculation were the channel length and the thickness of the film which were set equal to six different values, i.e., L=2, 4, 6, 8, 10, and 12μm and three different values, i.e., dsi =100, 200 and 300nm, respectively. The results indicate that the greatest photo-to-dark-current contrast ratio is achieved with L=10μm and dsi=200 nm at VGK=0.6V.
Abstract: In this paper, a comparative study of application of
supervised and unsupervised learning algorithms on illumination
invariant face recognition has been carried out. The supervised
learning has been carried out with the help of using a bi-layered
artificial neural network having one input, two hidden and one output
layer. The gradient descent with momentum and adaptive learning
rate back propagation learning algorithm has been used to implement
the supervised learning in a way that both the inputs and
corresponding outputs are provided at the time of training the
network, thus here is an inherent clustering and optimized learning of
weights which provide us with efficient results.. The unsupervised
learning has been implemented with the help of a modified
Counterpropagation network. The Counterpropagation network
involves the process of clustering followed by application of Outstar
rule to obtain the recognized face. The face recognition system has
been developed for recognizing faces which have varying
illumination intensities, where the database images vary in lighting
with respect to angle of illumination with horizontal and vertical
planes. The supervised and unsupervised learning algorithms have
been implemented and have been tested exhaustively, with and
without application of histogram equalization to get efficient results.
Abstract: In illumination variant face recognition, existing
methods extracting face albedo as light normalized image may lead to
loss of extensive facial details, with light template discarded. To
improve that, a novel approach for realistic facial texture
reconstruction by combining original image and albedo image is
proposed. First, light subspaces of different identities are established
from the given reference face images; then by projecting the original
and albedo image into each light subspace respectively, texture
reference images with corresponding lighting are reconstructed and
two texture subspaces are formed. According to the projections in
texture subspaces, facial texture with normal light can be synthesized.
Due to the combination of original image, facial details can be
preserved with face albedo. In addition, image partition is applied to
improve the synthesization performance. Experiments on Yale B and
CMUPIE databases demonstrate that this algorithm outperforms the
others both in image representation and in face recognition.
Abstract: In this paper a new approach to face recognition is presented that achieves double dimension reduction making the system computationally efficient with better recognition results. In pattern recognition techniques, discriminative information of image increases with increase in resolution to a certain extent, consequently face recognition results improve with increase in face image resolution and levels off when arriving at a certain resolution level. In the proposed model of face recognition, first image decimation algorithm is applied on face image for dimension reduction to a certain resolution level which provides best recognition results. Due to better computational speed and feature extraction potential of Discrete Cosine Transform (DCT) it is applied on face image. A subset of coefficients of DCT from low to mid frequencies that represent the face adequately and provides best recognition results is retained. A trade of between decimation factor, number of DCT coefficients retained and recognition rate with minimum computation is obtained. Preprocessing of the image is carried out to increase its robustness against variations in poses and illumination level. This new model has been tested on different databases which include ORL database, Yale database and a color database. The proposed technique has performed much better compared to other techniques. The significance of the model is two fold: (1) dimension reduction up to an effective and suitable face image resolution (2) appropriate DCT coefficients are retained to achieve best recognition results with varying image poses, intensity and illumination level.
Abstract: We fabricated the inverted-staggered etch stopper
structure oxide-based TFT and investigated the characteristics of oxide
TFT under the 400 nm wavelength light illumination. When 400 nm
light was illuminated, the threshold voltage (Vth) decreased and
subthreshold slope (SS) increased at forward sweep, while Vth and SS
were not altered when larger wavelength lights, such as 650 nm, 550
nm and 450 nm, were illuminated. At reverse sweep, the transfer curve
barely changed even under 400 nm light. Our experimental results
support that photo-induced hole carriers are captured by donor-like
interface trap and it caused the decrease of Vth and increase of SS. We
investigated the interface trap density increases proportionally to the
photo-induced hole concentration at active layer.
Abstract: Methods of clustering which were developed in the
data mining theory can be successfully applied to the investigation of
different kinds of dependencies between the conditions of
environment and human activities. It is known, that environmental
parameters such as temperature, relative humidity, atmospheric
pressure and illumination have significant effects on the human
mental performance. To investigate these parameters effect, data
mining technique of clustering using entropy and Information Gain
Ratio (IGR) K(Y/X) = (H(X)–H(Y/X))/H(Y) is used, where
H(Y)=-ΣPi ln(Pi). This technique allows adjusting the boundaries of
clusters. It is shown that the information gain ratio (IGR) grows
monotonically and simultaneously with degree of connectivity
between two variables. This approach has some preferences if
compared, for example, with correlation analysis due to relatively
smaller sensitivity to shape of functional dependencies. Variant of an
algorithm to implement the proposed method with some analysis of
above problem of environmental effects is also presented. It was
shown that proposed method converges with finite number of steps.
Abstract: Gabor-based face representation has achieved enormous success in face recognition. This paper addresses a novel algorithm for face recognition using neural networks trained by Gabor features. The system is commenced on convolving a face image with a series of Gabor filter coefficients at different scales and orientations. Two novel contributions of this paper are: scaling of rms contrast and introduction of fuzzily skewed filter. The neural network employed for face recognition is based on the multilayer perceptron (MLP) architecture with backpropagation algorithm and incorporates the convolution filter response of Gabor jet. The effectiveness of the algorithm has been justified over a face database with images captured at different illumination conditions.
Abstract: Colour choice has become a common strategy and
correlates highly with marketing. Three broad functions can be
identified for colour in a building context especially applied in
marketing communications, which are its role as an important
parameter in illumination designs, its capacity to influence the visual
appearance of a building in a predictable manner and as an aesthetic
function. The review of literatures shows that colour has an impact on
online marketing communications, and relations between colour, web
and marketing communications.
Abstract: Repeated observation of a given area over time yields
potential for many forms of change detection analysis. These
repeated observations are confounded in terms of radiometric
consistency due to changes in sensor calibration over time,
differences in illumination, observation angles and variation in
atmospheric effects.
This paper demonstrates applicability of an empirical relative
radiometric normalization method to a set of multitemporal cloudy
images acquired by Resourcesat1 LISS III sensor. Objective of this
study is to detect and remove cloud cover and normalize an image
radiometrically. Cloud detection is achieved by using Average
Brightness Threshold (ABT) algorithm. The detected cloud is
removed and replaced with data from another images of the same
area. After cloud removal, the proposed normalization method is
applied to reduce the radiometric influence caused by non surface
factors. This process identifies landscape elements whose reflectance
values are nearly constant over time, i.e. the subset of non-changing
pixels are identified using frequency based correlation technique. The
quality of radiometric normalization is statistically assessed by R2
value and mean square error (MSE) between each pair of analogous
band.
Abstract: One of the major cause of eye strain and other
problems caused while watching television is the relative illumination between the screen and its surrounding. This can be
overcome by adjusting the brightness of the screen with respect to the surrounding light. A controller based on fuzzy logic is proposed
in this paper. The fuzzy controller takes in the intensity of light
surrounding the screen and the present brightness of the screen as input. The output of the fuzzy controller is the grid voltage corresponding to the required brightness. This voltage is given to CRT and brightness is controller dynamically. For the given test system data, different de-fuzzifier methods have been implemented and the results are compared. In order to validate the effectiveness of
the proposed approach, a fuzzy controller has been designed by obtaining a test data from a real time system. The simulations are
performed in MATLAB and are verified with standard system data. The proposed approach can be implemented for real time
applications.