Abstract: System MEMORI automatically detects and recognizes rotated and/or rescaled versions of the objects of a database within digital color images with cluttered background. This task is accomplished by means of a region grouping algorithm guided by heuristic rules, whose parameters concern some geometrical properties and the recognition score of the database objects. This paper focuses on the strategies implemented in MEMORI for the estimation of the heuristic rule parameters. This estimation, being automatic, makes the system a highly user-friendly tool.
Abstract: Much research into handwritten Thai character
recognition have been proposed, such as comparing heads of
characters, Fuzzy logic and structure trees, etc. This paper presents a
system of handwritten Thai character recognition, which is based on
the Ant-minor algorithm (data mining based on Ant colony
optimization). Zoning is initially used to determine each character.
Then three distinct features (also called attributes) of each character
in each zone are extracted. The attributes are Head zone, End point,
and Feature code. All attributes are used for construct the
classification rules by an Ant-miner algorithm in order to classify
112 Thai characters. For this experiment, the Ant-miner algorithm is
adapted, with a small change to increase the recognition rate. The
result of this experiment is a 97% recognition rate of the training set
(11200 characters) and 82.7% recognition rate of unseen data test
(22400 characters).
Abstract: Results of Chilean wine classification based on the
information provided by an electronic nose are reported in this paper.
The classification scheme consists of two parts; in the first stage,
Principal Component Analysis is used as feature extraction method to
reduce the dimensionality of the original information. Then, Radial
Basis Functions Neural Networks is used as pattern recognition
technique to perform the classification. The objective of this study is
to classify different Cabernet Sauvignon, Merlot and Carménère wine
samples from different years, valleys and vineyards of Chile.
Abstract: This research work is aimed at speech recognition
using scaly neural networks. A small vocabulary of 11 words were
established first, these words are “word, file, open, print, exit, edit,
cut, copy, paste, doc1, doc2". These chosen words involved with
executing some computer functions such as opening a file, print
certain text document, cutting, copying, pasting, editing and exit.
It introduced to the computer then subjected to feature extraction
process using LPC (linear prediction coefficients). These features are
used as input to an artificial neural network in speaker dependent
mode. Half of the words are used for training the artificial neural
network and the other half are used for testing the system; those are
used for information retrieval.
The system components are consist of three parts, speech
processing and feature extraction, training and testing by using neural
networks and information retrieval.
The retrieve process proved to be 79.5-88% successful, which is
quite acceptable, considering the variation to surrounding, state of
the person, and the microphone type.
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: Existing literature ondesign reasoning seems to give
either one sided accounts on expert design behaviour based on
internal processing. In the same way ecological theoriesseem to
focus one sidedly on external elementsthat result in a lack of unifying
design cognition theory. Although current extended design cognition
studies acknowledge the intellectual interaction between internal and
external resources, there still seems to be insufficient understanding
of the complexities involved in such interactive processes. As
such,this paper proposes a novelmulti-directional model for design
researchers tomap the complex and dynamic conduct controlling
behaviour in which both the computational and ecological
perspectives are integrated in a vertical manner. A clear distinction
between identified intentional and emerging physical drivers, and
relationships between them during the early phases of experts- design
process, is demonstrated by presenting a case study in which the
model was employed.
Abstract: Building intelligent traffic guide systems has been an
interesting subject recently. A good system should be able to observe
all important visual information to be able to analyze the context of
the scene. To do so, signs in general, and traffic signs in particular,
are usually taken into account as they contain rich information to
these systems. Therefore, many researchers have put an effort on
sign recognition field. Sign localization or sign detection is the most
important step in the sign recognition process. This step filters out
non informative area in the scene, and locates candidates in later
steps. In this paper, we apply a new approach in detecting sign
locations using a new color invariant model. Experiments are carried
out with different datasets introduced in other works where authors
claimed the difficulty in detecting signs under unfavorable imaging
conditions. Our method is simple, fast and most importantly it gives
a high detection rate in locating signs.
Abstract: The practical implementation of audio-video coupled speech recognition systems is mainly limited by the hardware complexity to integrate two radically different information capturing devices with good temporal synchronisation. In this paper, we propose a solution based on a smart CMOS image sensor in order to simplify the hardware integration difficulties. By using on-chip image processing, this smart sensor can calculate in real time the X/Y projections of the captured image. This on-chip projection reduces considerably the volume of the output data. This data-volume reduction permits a transmission of the condensed visual information via the same audio channel by using a stereophonic input available on most of the standard computation devices such as PC, PDA and mobile phones. A prototype called VMIKE (Visio-Microphone) has been designed and realised by using standard 0.35um CMOS technology. A preliminary experiment gives encouraged results. Its efficiency will be further investigated in a large variety of applications such as biometrics, speech recognition in noisy environments, and vocal control for military or disabled persons, etc.
Abstract: This work presents a neural network model for the
clustering analysis of data based on Self Organizing Maps (SOM).
The model evolves during the training stage towards a hierarchical
structure according to the input requirements. The hierarchical structure
symbolizes a specialization tool that provides refinements of the
classification process. The structure behaves like a single map with
different resolutions depending on the region to analyze. The benefits
and performance of the algorithm are discussed in application to the
Iris dataset, a classical example for pattern recognition.
Abstract: Skin color is an important visual cue for computer
vision systems involving human users. In this paper we combine skin
color and optical flow for detection and tracking of skin regions. We
apply these techniques to gesture recognition with encouraging
results. We propose a novel skin similarity measure. For grouping
detected skin regions we propose a novel skin region grouping
mechanism. The proposed techniques work with any number of skin
regions making them suitable for a multiuser scenario.
Abstract: the cursive nature of the Arabic writing makes it
difficult to accurately segment characters or even deal with the whole
word efficiently. Therefore, in this paper, a printed Arabic sub-word
recognition system is proposed. The suggested algorithm utilizes
geometrical moments as descriptors for the separated sub-words.
Three types of moments are investigated and applied to the printed
sub-word images after dividing each image into multiple parts using
windowing. Since moments are global descriptors, the windowing
mechanism allows the moments to be applied to local regions of the
sub-word. The local-global mixture of the proposed scheme increases
the discrimination power of the moments while keeping the
simplicity and ease of use of moments.
Abstract: Identity verification of authentic persons by their multiview faces is a real valued problem in machine vision. Multiview faces are having difficulties due to non-linear representation in the feature space. This paper illustrates the usability of the generalization of LDA in the form of canonical covariate for face recognition to multiview faces. In the proposed work, the Gabor filter bank is used to extract facial features that characterized by spatial frequency, spatial locality and orientation. Gabor face representation captures substantial amount of variations of the face instances that often occurs due to illumination, pose and facial expression changes. Convolution of Gabor filter bank to face images of rotated profile views produce Gabor faces with high dimensional features vectors. Canonical covariate is then used to Gabor faces to reduce the high dimensional feature spaces into low dimensional subspaces. Finally, support vector machines are trained with canonical sub-spaces that contain reduced set of features and perform recognition task. The proposed system is evaluated with UMIST face database. The experiment results demonstrate the efficiency and robustness of the proposed system with high recognition rates.
Abstract: This research paper deals with the implementation of face recognition using neural network (recognition classifier) on low-resolution images. The proposed system contains two parts, preprocessing and face classification. The preprocessing part converts original images into blurry image using average filter and equalizes the histogram of those image (lighting normalization). The bi-cubic interpolation function is applied onto equalized image to get resized image. The resized image is actually low-resolution image providing faster processing for training and testing. The preprocessed image becomes the input to neural network classifier, which uses back-propagation algorithm to recognize the familiar faces. The crux of proposed algorithm is its beauty to use single neural network as classifier, which produces straightforward approach towards face recognition. The single neural network consists of three layers with Log sigmoid, Hyperbolic tangent sigmoid and Linear transfer function respectively. The training function, which is incorporated in our work, is Gradient descent with momentum (adaptive learning rate) back propagation. The proposed algorithm was trained on ORL (Olivetti Research Laboratory) database with 5 training images. The empirical results provide the accuracy of 94.50%, 93.00% and 90.25% for 20, 30 and 40 subjects respectively, with time delay of 0.0934 sec per image.
Abstract: Because of increasing demands for security in today-s
society and also due to paying much more attention to machine
vision, biometric researches, pattern recognition and data retrieval in
color images, face detection has got more application. In this article
we present a scientific approach for modeling human skin color, and
also offer an algorithm that tries to detect faces within color images
by combination of skin features and determined threshold in the
model. Proposed model is based on statistical data in different color
spaces. Offered algorithm, using some specified color threshold, first,
divides image pixels into two groups: skin pixel group and non-skin
pixel group and then based on some geometric features of face
decides which area belongs to face.
Two main results that we received from this research are as follow:
first, proposed model can be applied easily on different databases and
color spaces to establish proper threshold. Second, our algorithm can
adapt itself with runtime condition and its results demonstrate
desirable progress in comparison with similar cases.
Abstract: The Artificial immune systems algorithms are Meta
heuristic optimization method, which are used for clustering and
pattern recognition applications are abundantly. These algorithms in
multimodal optimization problems are more efficient than genetic
algorithms. A major drawback in these algorithms is their slow
convergence to global optimum and their weak stability can be
considered in various running of these algorithms. In this paper,
improved Artificial Immune System Algorithm is introduced for the
first time to overcome its problems of artificial immune system. That
use of the small size of a local search around the memory antibodies
is used for improving the algorithm efficiently. The credibility of the
proposed approach is evaluated by simulations, and it is shown that
the proposed approach achieves better results can be achieved
compared to the standard artificial immune system algorithms
Abstract: In this work a new offline signature recognition system
based on Radon Transform, Fractal Dimension (FD) and Support Vector Machine (SVM) is presented. In the first step, projections of
original signatures along four specified directions have been performed using radon transform. Then, FDs of four obtained
vectors are calculated to construct a feature vector for each
signature. These vectors are then fed into SVM classifier for recognition of signatures. In order to evaluate the effectiveness of
the system several experiments are carried out. Offline signature
database from signature verification competition (SVC) 2004 is used
during all of the tests. Experimental result indicates that the proposed method achieved high accuracy rate in signature recognition.
Abstract: One of the popular methods for recognition of facial
expressions such as happiness, sadness and surprise is based on
deformation of facial features. Motion vectors which show these
deformations can be specified by the optical flow. In this method, for
detecting emotions, the resulted set of motion vectors are compared
with standard deformation template that caused by facial expressions.
In this paper, a new method is introduced to compute the quantity of
likeness in order to make decision based on the importance of
obtained vectors from an optical flow approach. For finding the
vectors, one of the efficient optical flow method developed by
Gautama and VanHulle[17] is used. The suggested method has been
examined over Cohn-Kanade AU-Coded Facial Expression Database,
one of the most comprehensive collections of test images available.
The experimental results show that our method could correctly
recognize the facial expressions in 94% of case studies. The results
also show that only a few number of image frames (three frames) are
sufficient to detect facial expressions with rate of success of about
83.3%. This is a significant improvement over the available methods.
Abstract: Using spatial models as a shared common basis of
information about the environment for different kinds of contextaware
systems has been a heavily researched topic in the last years.
Thereby the research focused on how to create, to update, and to
merge spatial models so as to enable highly dynamic, consistent and
coherent spatial models at large scale. In this paper however, we
want to concentrate on how context-aware applications could use this
information so as to adapt their behavior according to the situation
they are in. The main idea is to provide the spatial model
infrastructure with a situation recognition component based on
generic situation templates. A situation template is – as part of a
much larger situation template library – an abstract, machinereadable
description of a certain basic situation type, which could be
used by different applications to evaluate their situation. In this
paper, different theoretical and practical issues – technical, ethical
and philosophical ones – are discussed important for understanding
and developing situation dependent systems based on situation
templates. A basic system design is presented which allows for the
reasoning with uncertain data using an improved version of a
learning algorithm for the automatic adaption of situation templates.
Finally, for supporting the development of adaptive applications, we
present a new situation-aware adaptation concept based on
workflows.
Abstract: This paper compares Hilditch, Rosenfeld, Zhang-
Suen, dan Nagendraprasad Wang Gupta (NWG) thinning algorithms
for Javanese character image recognition. Thinning is an effective
process when the focus in not on the size of the pattern, but rather on
the relative position of the strokes in the pattern. The research
analyzes the thinning of 60 Javanese characters.
Time-wise, Zhang-Suen algorithm gives the best results with the
average process time being 0.00455188 seconds. But if we look at
the percentage of pixels that meet one-pixel thickness, Rosenfelt
algorithm gives the best results, with a 99.98% success rate. From the
number of pixels that are erased, NWG algorithm gives the best
results with the average number of pixels erased being 84.12%. It can
be concluded that the Hilditch algorithm performs least successfully
compared to the other three algorithms.
Abstract: This paper presents an approach based on the
adoption of a distributed cognition framework and a non parametric
multicriteria evaluation methodology (DEA) designed specifically to
compare e-commerce websites from the consumer/user viewpoint. In
particular, the framework considers a website relative efficiency as a
measure of its quality and usability. A website is modelled as a black
box capable to provide the consumer/user with a set of
functionalities. When the consumer/user interacts with the website to
perform a task, he/she is involved in a cognitive activity, sustaining a
cognitive cost to search, interpret and process information, and
experiencing a sense of satisfaction. The degree of ambiguity and
uncertainty he/she perceives and the needed search time determine
the effort size – and, henceforth, the cognitive cost amount – he/she
has to sustain to perform his/her task. On the contrary, task
performing and result achievement induce a sense of gratification,
satisfaction and usefulness. In total, 9 variables are measured,
classified in a set of 3 website macro-dimensions (user experience,
site navigability and structure). The framework is implemented to
compare 40 websites of businesses performing electronic commerce
in the information technology market. A questionnaire to collect
subjective judgements for the websites in the sample was purposely
designed and administered to 85 university students enrolled in
computer science and information systems engineering
undergraduate courses.